CN115902004A - Measuring device and measuring method for conductivity of degassed hydrogen - Google Patents
Measuring device and measuring method for conductivity of degassed hydrogen Download PDFInfo
- Publication number
- CN115902004A CN115902004A CN202211388392.4A CN202211388392A CN115902004A CN 115902004 A CN115902004 A CN 115902004A CN 202211388392 A CN202211388392 A CN 202211388392A CN 115902004 A CN115902004 A CN 115902004A
- Authority
- CN
- China
- Prior art keywords
- map
- feature
- hydrogen conductivity
- value
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 229910052739 hydrogen Inorganic materials 0.000 title claims abstract description 175
- 239000001257 hydrogen Substances 0.000 title claims abstract description 175
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 title claims abstract description 166
- 238000000034 method Methods 0.000 title claims abstract description 22
- 239000007788 liquid Substances 0.000 claims abstract description 118
- 238000005259 measurement Methods 0.000 claims abstract description 79
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 claims abstract description 78
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 70
- 229910002092 carbon dioxide Inorganic materials 0.000 claims abstract description 39
- 239000001569 carbon dioxide Substances 0.000 claims abstract description 33
- 238000010586 diagram Methods 0.000 claims description 56
- 238000004811 liquid chromatography Methods 0.000 claims description 54
- 239000011159 matrix material Substances 0.000 claims description 30
- 238000013527 convolutional neural network Methods 0.000 claims description 24
- 230000004913 activation Effects 0.000 claims description 19
- 230000004927 fusion Effects 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 13
- 238000000605 extraction Methods 0.000 claims description 11
- 238000012937 correction Methods 0.000 claims description 10
- 238000011176 pooling Methods 0.000 claims description 10
- 150000002431 hydrogen Chemical class 0.000 claims description 9
- 238000003062 neural network model Methods 0.000 claims description 9
- 238000007872 degassing Methods 0.000 claims description 5
- 230000008447 perception Effects 0.000 claims description 4
- BVKZGUZCCUSVTD-UHFFFAOYSA-L Carbonate Chemical compound [O-]C([O-])=O BVKZGUZCCUSVTD-UHFFFAOYSA-L 0.000 abstract description 19
- 230000008901 benefit Effects 0.000 abstract description 14
- 230000007774 longterm Effects 0.000 abstract description 12
- 238000001514 detection method Methods 0.000 abstract description 6
- 230000007246 mechanism Effects 0.000 description 8
- 239000000243 solution Substances 0.000 description 8
- 238000013135 deep learning Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000002708 enhancing effect Effects 0.000 description 4
- 239000007789 gas Substances 0.000 description 4
- BVKZGUZCCUSVTD-UHFFFAOYSA-M Bicarbonate Chemical compound OC([O-])=O BVKZGUZCCUSVTD-UHFFFAOYSA-M 0.000 description 3
- QAOWNCQODCNURD-UHFFFAOYSA-L Sulfate Chemical compound [O-]S([O-])(=O)=O QAOWNCQODCNURD-UHFFFAOYSA-L 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 239000012535 impurity Substances 0.000 description 3
- 238000005065 mining Methods 0.000 description 3
- 239000002245 particle Substances 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- VEXZGXHMUGYJMC-UHFFFAOYSA-M Chloride anion Chemical compound [Cl-] VEXZGXHMUGYJMC-UHFFFAOYSA-M 0.000 description 2
- 238000010790 dilution Methods 0.000 description 2
- 239000012895 dilution Substances 0.000 description 2
- 150000002500 ions Chemical class 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000007789 sealing Methods 0.000 description 2
- VEXZGXHMUGYJMC-UHFFFAOYSA-N Hydrochloric acid Chemical compound Cl VEXZGXHMUGYJMC-UHFFFAOYSA-N 0.000 description 1
- 150000001450 anions Chemical class 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000006356 dehydrogenation reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007062 hydrolysis Effects 0.000 description 1
- 238000006460 hydrolysis reaction Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Landscapes
- Investigating Or Analyzing Materials By The Use Of Electric Means (AREA)
Abstract
The application relates to the field of intelligent detection, and particularly discloses a measuring device and a measuring method for degassed hydrogen conductivity, wherein a degassed hydrogen conductivity measured value of condensed water affected by carbon dioxide is obtained through measurement of a hydrogen conductivity meter, and implicit characteristic information of the content and characteristics of carbonate ions is extracted from a liquid chromatogram of the condensed water, so that decoding is carried out to obtain a degassed hydrogen conductivity affected value of the carbon dioxide. Further, the difference between the two is used to represent the true value of the degassed hydrogen conductivity measurement of the condensed water. Thus, the true value of the degassed hydrogen conductivity of the condensed water can be accurately measured to ensure the safety of the equipment and the long-term economic benefit.
Description
Technical Field
The present application relates to the field of intelligent detection, and more particularly, to a measuring apparatus and a measuring method for degassed hydrogen conductivity.
Background
The gas-steam combined cycle generator set operates in a mode of 'early start and late stop' peak regulation, for energy conservation, after the generator set is generally stopped, the vacuum pump and the condensate pump are stopped, a large amount of air enters the condenser due to low self-sealing water pressure of the condensate pump, the air cannot be pumped out in time due to the stop of the vacuum pump, the content of dissolved gas in the condensate water is increased, part of CO2 in the dissolved gas can be hydrolyzed to generate carbonate and bicarbonate, and the hydrogen conductivity of the condensate water is greatly increased.
From the viewpoint of equipment safety and long-term economic benefit, the research on the operation monitoring parameters of the low-pressure cylinder of the steam turbine has extremely important significance. Meanwhile, condensed water in a steam-water system of a thermal power plant contains a certain amount of CO2, and the CO2 is mixed with chloride ions and sulfate ions, and the content of negative ion impurities which cannot be directly reflected is measured by using a common hydrogen conductivity meter, so that misjudgment of hydrogen conductivity can be caused.
Therefore, an optimized measuring device for the degassed hydrogen conductivity is desired.
Disclosure of Invention
The present application is proposed to solve the above-mentioned technical problems. The embodiment of the application provides a measuring device and a measuring method for degassed hydrogen conductivity, which are used for obtaining a degassed hydrogen conductivity measured value of condensed water affected by carbon dioxide through hydrogen conductivity meter measurement, and extracting implicit characteristic information of content and characteristics of carbonate ions from a liquid chromatogram of the condensed water so as to decode the extracted information to obtain a degassed hydrogen conductivity affected value of the carbon dioxide. Further, the difference between the two is used to represent the true value of the degassed hydrogen conductivity measurement of the condensed water. Thus, the true value of the degassed hydrogen conductivity of the condensed water can be accurately measured to ensure the safety of the equipment and the long-term economic benefit.
According to an aspect of the present application, there is provided a measuring apparatus of degassed hydrogen conductivity, including:
the common measurement module is used for obtaining a degassed hydrogen conductivity measurement value of the condensate water to be detected through a common hydrogen conductivity meter;
the liquid chromatogram acquisition module is used for acquiring a liquid chromatogram of the condensate to be detected;
the liquid chromatogram characteristic extraction module is used for enabling the liquid chromatogram of the condensate water to be detected to pass through a deep convolution neural network model serving as a characteristic extractor to obtain a liquid chromatogram characteristic diagram;
the chromatogram characteristic enhancement module is used for enabling the liquid chromatogram characteristic diagram to pass through the parallel weight distribution module to obtain an enhanced liquid chromatogram characteristic diagram;
the decoding module is used for enabling the enhanced liquid chromatography characteristic diagram to pass through a decoder to obtain a decoded value used for representing a degassed hydrogen conductivity influence value of the carbon dioxide; and
and the real measurement value generating module is used for calculating the difference value between the measured value of the degassed hydrogen conductivity and the decoded value to obtain the real value of the degassed hydrogen conductivity measurement.
In the above apparatus for measuring a conductivity of degassed hydrogen, the liquid chromatogram feature extraction module is further configured to: performing, in a layer forward pass, input data using the layers of the deep convolutional neural network model as a feature extractor: performing convolution processing on input data to obtain a convolution characteristic diagram; performing pooling processing based on a local feature matrix on the convolution feature map to obtain a pooled feature map; and performing nonlinear activation on the pooled feature map to obtain an activated feature map; the output of the last layer of the deep convolutional neural network serving as the feature extractor is the liquid chromatogram feature map, and the input of the first layer of the deep convolutional neural network serving as the feature extractor is the liquid chromatogram of the condensate water to be detected.
In the above apparatus for measuring a conductivity of degassed hydrogen, the chromatogram feature enhancing module includes: a spatial attention unit for inputting the liquid chromatography profile into a spatial attention module of the parallel weight assignment module to obtain a spatial attention map; a channel attention unit for inputting the liquid chromatography feature map into a channel attention module of the parallel weight assignment module to obtain a channel attention map; and a fusion unit for fusing the spatial attention map and the channel attention map to obtain the enhanced liquid chromatography characteristic map.
In the above apparatus for measuring a conductivity of dehydrated hydrogen, the spatial attention unit includes: the spatial perception subunit is used for carrying out convolutional coding on the liquid chromatogram characteristic diagram by using the convolutional layer of the spatial attention module to obtain a spatial attention matrix; a probabilistic subunit, configured to input the spatial attention moment array into a Softmax activation function of the spatial attention module to obtain a spatial attention score map; and a spatial attention applying subunit, configured to calculate a product of the spatial attention score map and the liquid chromatogram feature map by position points to obtain the spatial attention map.
In the above apparatus for measuring a conductivity of degassed hydrogen, the channel attention unit includes: the global mean pooling subunit is used for calculating the global mean of each feature matrix of the liquid chromatogram feature map along the channel dimension to obtain a channel feature vector; a nonlinear activation subunit, configured to input the channel feature vector into a Softmax activation function to obtain a channel attention weight feature vector; and the channel attention applying subunit is used for respectively weighting each feature matrix of the liquid chromatogram feature map along the channel dimension by taking the feature value of each position in the channel attention weight feature vector as a weight so as to obtain the channel attention map.
In the above apparatus for measuring a conductivity of degassed hydrogen, the fusion unit includes: a corrector subunit, configured to perform feature correction on the spatial attention map and the channel attention map respectively to obtain a corrected spatial attention map and a corrected channel attention map according to the following formulas:
wherein
And &>
Is a characteristic value for the respective position of the spatial attention map and of the channel attention map, and->
And &>
Is the mean of all the characteristic values of the spatial attention map and the channel attention map, respectively, log represents the base-2 logarithmic function value; and a position-wise fusion subunit for calculating a position-wise sum between the corrected spatial attention map and the corrected channel attention map to obtain the enhanced liquid chromatography feature map.
In the above apparatus for measuring a conductivity of degassed hydrogen, the decoding module is further configured to: performing decoding regression on the enhanced liquid chromatography characteristic diagram by using the decoder according to the following formula to obtain a decoded value; wherein the formula is:
wherein X represents the enhanced liquid chromatography profile, Y is the decoded value, W is a weight matrix, and>
representing a matrix multiplication.
According to another aspect of the present application, there is provided a method of measuring a degassed hydrogen conductivity, comprising:
obtaining a measured value of the degassed hydrogen conductivity of the condensate to be detected through a common hydrogen conductivity meter;
acquiring a liquid chromatogram of the condensate to be detected;
the liquid chromatogram of the condensed water to be detected is processed by a deep convolution neural network model as a feature extractor to obtain a liquid chromatogram feature map;
passing the liquid chromatography profile through a parallel weight assignment module to obtain an enhanced liquid chromatography profile;
passing the enhanced liquid chromatography characteristic map through a decoder to obtain decoded values representing the influence values of the conductivity of the degassed hydrogen of carbon dioxide; and
and calculating the difference value between the measured value of the conductivity of the degassed hydrogen and the decoded value to obtain a true value of the measured conductivity of the degassed hydrogen.
Compared with the prior art, the measuring device and the measuring method for the conductivity of the degassed hydrogen provided by the application have the advantages that the degassed hydrogen conductivity measured value of the condensed water affected by carbon dioxide is obtained through the hydrogen conductivity meter measurement, and the implicit characteristic information of the content and the characteristics of carbonate ions is extracted from the liquid chromatogram of the condensed water, so that the degassed hydrogen conductivity affected value of the carbon dioxide is obtained through decoding. Further, the difference between the two is used to represent the true value of the degassed hydrogen conductivity measurement of the condensed water. Thus, the real value of the degassed hydrogen conductivity of the condensed water can be accurately measured to ensure the safety of the equipment and the long-term economic benefit.
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 indicate like parts or steps.
Fig. 1 is a view of an application scenario of a measuring apparatus for degassed hydrogen conductivity according to an embodiment of the present application;
fig. 2 is a block diagram of a degassed hydrogen conductivity measurement device according to an embodiment of the application;
FIG. 3 is a system architecture diagram of a degassed hydrogen conductivity measurement device according to an embodiment of the present application;
FIG. 4 is a flow chart of convolutional neural network coding in a degassed hydrogen conductivity measurement device according to an embodiment of the present application;
FIG. 5 is a block diagram of a chromatogram feature enhancement module in a measuring apparatus of degassed hydrogen conductivity according to an embodiment of the present application;
FIG. 6 is a block diagram of a spatial attention cell in a degassed hydrogen conductivity measurement device according to an embodiment of the present application;
fig. 7 is a block diagram of a fusion unit in a measuring apparatus of degassed hydrogen conductivity according to an embodiment of the application;
fig. 8 is a flowchart of a method of measuring degassed hydrogen conductivity according to an embodiment of the 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, the gas-steam combined cycle generator set operates in a mode of peak shaving operation in a mode of 'early start and late stop', for the requirement of energy saving, after the unit is stopped, the vacuum pump and the condensate pump are usually stopped, as the self-sealing water pressure of the condensate pump is low, a large amount of air enters the condenser, the vacuum pump is stopped, so that the part of air cannot be pumped out in time, the content of dissolved gas in the condensate is increased, part of CO2 in the dissolved gas can be hydrolyzed to generate carbonate and bicarbonate, and the hydrogen conductivity of the condensate is greatly increased.
From the viewpoint of equipment safety and long-term economic benefit, the research on the operation monitoring parameters of the low-pressure cylinder of the steam turbine has extremely important significance. Meanwhile, condensed water in a steam-water system of a thermal power plant contains a certain amount of CO2, and the CO2 is mixed with chloride ions and sulfate ions, and the content of negative ion impurities which cannot be directly reflected is measured by using a common hydrogen conductivity meter, so that misjudgment of hydrogen conductivity can be caused. Therefore, an optimized measuring device for the degassed hydrogen conductivity is desired.
At present, deep learning and neural networks have been widely applied in the fields of computer vision, natural language processing, speech signal processing, and the like. In addition, deep learning and neural networks also exhibit a level close to or even exceeding that of humans in the fields of image classification, object detection, semantic segmentation, text translation, and the like.
In recent years, deep learning and the development of neural networks have provided new solutions and solutions for the measurement of the electrical conductivity of degassed hydrogen.
Accordingly, it is considered that in the measurement of the degassed hydrogen conductivity, the hydrogen conductivity of the condensed water greatly rises due to the hydrolysis of carbon dioxide to generate carbonate and bicarbonate, and it is confused with chloride and sulfate ions to cause the measurement of the content of the anion impurity which cannot be directly reacted using the general hydrogen conductivity meter. Based on this, in the technical solution of the present application, it is desirable to adopt an artificial intelligence detection technology based on deep learning, obtain a degassed hydrogen conductivity measurement value of the condensed water affected by carbon dioxide through hydrogen conductivity meter measurement, and extract implicit characteristic information of the content and characteristics of carbonate ions from a liquid chromatogram of the condensed water, so as to perform decoding to obtain a degassed hydrogen conductivity affected value of carbon dioxide. Further, the difference between the two is used to represent the true value of the degassed hydrogen conductivity measurement of the condensed water. Thus, the real value of the degassed hydrogen conductivity of the condensed water can be accurately measured to ensure the safety of the equipment and the long-term economic benefit.
Specifically, in the technical solution of the present application, first, a degassed hydrogen conductivity measurement value of the condensate to be detected is obtained through a common hydrogen conductivity meter. Then, considering that the content and the characteristics of carbonate particles can be observed from the liquid chromatogram, a liquid chromatogram of the condensate to be detected is further acquired, and the obtained liquid chromatogram of the condensate to be detected is subjected to feature mining. Specifically, hidden feature mining is performed on the liquid chromatogram of the condensate to be detected through a deep convolutional neural network model serving as a feature extractor, so as to extract feature distribution representation of local hidden features in a high-dimensional space in the liquid chromatogram of the condensate to be detected, namely hidden distribution feature information about carbonate ions in the liquid chromatogram of the condensate to be detected, and thus a liquid chromatogram feature map is obtained.
It will be appreciated that in measuring the degassed hydrogen conductivity influence value of carbon dioxide, the content characteristics and characteristic characteristics of the carbonate ions therein, i.e. the hidden characteristics of the hidden distribution characteristics of the carbonate ions in spatial position and channel dimension, should be focused on while ignoring unwanted interference characteristics not related to the degassed hydrogen conductivity influence value measurement of carbon dioxide. Therefore, in the technical solution of the present application, in order to solve the problem that the target detection accuracy is low due to the edge blurring in the liquid chromatography feature map, a parallel weight assignment module is used to perform feature enhancement in the liquid chromatography feature map. Specifically, the liquid chromatogram characteristic diagram is processed by a parallel weight distribution module to obtain an enhanced liquid chromatogram characteristic diagram, so that effective characteristic representation can be enhanced, useless characteristic information can be suppressed, and the accuracy of subsequent classification can be improved. In particular, the parallel weight assignment module performs feature enhancement on the liquid chromatogram feature map by using a spatial attention module and a channel attention module respectively, wherein the extracted image features of the channel attention reflect the correlation and importance among feature channels, and the extracted image features of the spatial attention reflect the weight of feature difference in spatial dimension, so as to suppress or enhance features of different spatial positions.
Then, the enhanced liquid chromatography characteristic diagram is used as a decoding characteristic diagram to carry out decoding regression through a decoder so as to obtain a decoding value for representing the degassed hydrogen conductivity influence value of the carbon dioxide. That is, the content and characteristics of carbonate particles can be observed from the liquid chromatogram of the condensed water, and thus, a decoded value for representing an influence value of the degassed hydrogen conductivity of carbon dioxide is obtained based on the structure of the feature extractor + decoder. And then calculating the difference value between the measured value of the degassed hydrogen conductivity and the decoded value to obtain a true value of the degassed hydrogen conductivity measurement. In this way, the true value of the degassed hydrogen conductivity of the condensed water can be measured.
In particular, in the solution of the present application, the parallel weight assignment module derives from the liquid chromatography profile F 1 Obtaining a spatial attention diagram F by a spatial attention mechanism and a channel attention mechanism respectively 2 And channel attention diagram F 3 However, since the spatial attention mechanism and the channel attention mechanism respectively enhance feature extraction in different dimensions, this enables the spatial attention scheme F 2 And the channel attention map F 3 The overall feature distribution has a spatial position error in a high-dimensional feature space, which affects the fusion of the spatial attention diagrams F by means of point addition 2 And the channel attention map F 3 The fusion effect of (2).
In addition, the applicant of the present application considered the spatial attention map F 2 And the channel attention map F 3 Relative to the liquid chromatogram profile F 1 Are homologous and thus have a correspondence such that the spatial attention map F can be applied 2 And the channel attention map F 3 The relative angle probability information representation correction is respectively carried out, and the representation is expressed as follows:
wherein
And &>
Respectively the spatial attention map F 2 And the channel attention map F 3 Is determined by the characteristic value of (a), and->
And &>
Is the spatial attention map F 2 And the channel attention map F 3 Is calculated as the mean of all characteristic values of (1).
Here, the relative angle-like probability information indicates that the correction is performed by the spatial attention map F 2 And the channel attention map F 3 Relative class angle probability information representation between to perform the space attention diagram F 2 And the channel attention map F 3 Geometric dilution of spatial position error of feature distribution in high-dimensional feature space, thereby performing F-shaped attention in the space 2 And the channel attention map F 3 Has a certain corresponding relation, based on the space attention diagram F 2 And the channel attention map F 3 The spatial attention map F is improved by performing implicit context correspondence correction of features by point-by-point regression of positions in comparison with the distribution constraint of the respective positions as a whole 2 And the channel attention map F 3 And the accuracy of decoding regression is improved by the fusion effect of point-and-add mode fusion. Thus, the real value of the degassed hydrogen conductivity of the condensed water can be accurately measured to ensure the safety of the equipment and the long-term economic benefit.
Based on this, the present application proposes a measuring device of degassed hydrogen conductivity, comprising: the common measurement module is used for obtaining a degassed hydrogen conductivity measurement value of the condensate to be detected through a common hydrogen conductivity meter; the liquid chromatogram acquisition module is used for acquiring a liquid chromatogram of the condensate to be detected; the liquid chromatogram characteristic extraction module is used for enabling the liquid chromatogram of the condensate water to be detected to pass through a deep convolution neural network model serving as a characteristic extractor to obtain a liquid chromatogram characteristic diagram; the chromatogram characteristic enhancement module is used for enabling the liquid chromatogram characteristic graph to pass through the parallel weight distribution module to obtain an enhanced liquid chromatogram characteristic graph; the decoding module is used for enabling the enhanced liquid chromatography characteristic diagram to pass through a decoder to obtain a decoded value used for representing a degassed hydrogen conductivity influence value of the carbon dioxide; and the measurement true value generation module is used for calculating the difference value between the degassing hydrogen conductivity measurement value and the decoding value to obtain the degassing hydrogen conductivity measurement true value.
Fig. 1 is a view of an application scenario of a measuring apparatus for degassed hydrogen conductivity according to an embodiment of the present application. As shown in fig. 1, in this application scenario, a degassed hydrogen conductivity measurement of the condensate to be detected is taken via a common hydrogen conductivity meter (e.g., H as illustrated in fig. 1); and acquiring a liquid chromatogram (e.g., F as illustrated in fig. 1) of the condensate to be detected by a chromatograph (e.g., C as illustrated in fig. 1). Then, the above information is inputted into a server (for example, S in fig. 1) deployed with a measurement algorithm for degassed hydrogen conductivity, wherein the server can process the above inputted information with the measurement algorithm for degassed hydrogen conductivity to generate a decoded value representing a degassed hydrogen conductivity influence value of carbon dioxide, and further calculate a difference value between the measured value of degassed hydrogen conductivity and the decoded value to obtain a true value of degassed hydrogen conductivity measurement.
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 is a block diagram of a measuring apparatus of degassed hydrogen conductivity according to an embodiment of the present application. As shown in fig. 2, the degassed hydrogen conductivity measurement device 300 according to the embodiment of the present application includes: a common measurement module 310; a liquid chromatogram acquisition module 320; a liquid chromatogram feature extraction module 330; a chromatogram feature enhancement module 340; a decoding module 350; and a measured true value generation module 360.
The common measurement module 310 is configured to obtain a degassed hydrogen conductivity measurement value of the condensed water to be detected through a common hydrogen conductivity meter; the liquid chromatogram acquisition module 320 is used for acquiring a liquid chromatogram of the to-be-detected condensed water; the liquid chromatogram feature extraction module 330 is configured to pass the liquid chromatogram of the to-be-detected condensed water through a deep convolution neural network model serving as a feature extractor to obtain a liquid chromatogram feature map; the chromatogram feature enhancement module 340 is configured to pass the liquid chromatogram feature through a parallel weight distribution module to obtain an enhanced liquid chromatogram feature; the decoding module 350, configured to pass the enhanced liquid chromatography characteristic map through a decoder to obtain a decoded value representing a degassed hydrogen conductivity influence value of carbon dioxide; and the real measurement value generation module 360 is configured to calculate a difference between the measured value of the degassed hydrogen conductivity and the decoded value to obtain a real value of the degassed hydrogen conductivity measurement.
Fig. 3 is a system architecture diagram of a measuring apparatus of degassed hydrogen conductivity according to an embodiment of the application. As shown in fig. 3, in the system architecture of the measuring apparatus 300 for degassed hydrogen conductivity, firstly, a degassed hydrogen conductivity measurement value of the condensed water to be detected is obtained through a common hydrogen conductivity meter by the common measurement module 310; then, the liquid chromatogram acquisition module 320 acquires a liquid chromatogram of the to-be-detected condensed water; the liquid chromatogram characteristic extraction module 330 is used for obtaining a liquid chromatogram characteristic diagram by using the liquid chromatogram of the to-be-detected condensed water acquired by the liquid chromatogram acquisition module 320 through a deep convolution neural network model serving as a characteristic extractor; then, the chromatogram feature enhancing module 340 obtains an enhanced liquid chromatogram feature map by passing the liquid chromatogram feature map obtained by the liquid chromatogram feature extracting module 330 through a parallel weight distribution module; the decoding module 350 passes the enhanced liquid chromatography characteristic map obtained by the chromatogram characteristic enhancing module 340 through a decoder to obtain a decoded value representing the influence value of the degassed hydrogen conductivity of the carbon dioxide; further, the real measurement value generation module 360 calculates a difference between the measured value of the degassed hydrogen conductivity and the decoded value to obtain a real value of the degassed hydrogen conductivity measurement.
Specifically, in the operation process of the degassed hydrogen conductivity measurement device 300, the common measurement module 310 and the liquid chromatogram acquisition module 320 are configured to obtain a degassed hydrogen conductivity measurement value of the to-be-detected condensed water through a common hydrogen conductivity meter, and then obtain a liquid chromatogram of the to-be-detected condensed water. In a specific example of the application, a chromatograph is used to obtain a liquid chromatogram of the condensed water to be detected, and implicit characteristic information of the content and the characteristics of the carbonate ions is extracted from the liquid chromatogram of the condensed water, so as to decode the liquid chromatogram to obtain the degassed hydrogen conductivity influence value of the carbon dioxide. Further, the difference between the two is used to represent the true value of the degassed hydrogen conductivity measurement of the condensed water. Thus, the true value of the degassed hydrogen conductivity of the condensed water can be accurately measured to ensure the safety of the equipment and the long-term economic benefit.
Specifically, during the operation of the degassed hydrogen conductivity measurement apparatus 300, the liquid chromatogram feature extraction module 330 is configured to pass the liquid chromatogram of the to-be-detected condensed water through a deep convolution neural network model as a feature extractor to obtain a liquid chromatogram feature map. In the technical scheme of the application, implicit feature mining is performed on the liquid chromatogram of the to-be-detected condensed water through a deep convolutional neural network model serving as a feature extractor, so as to extract feature distribution representation of local implicit features in the liquid chromatogram of the to-be-detected condensed water in a high-dimensional space, namely hidden distribution feature information about carbonate ions in the liquid chromatogram of the to-be-detected condensed water, and thus a liquid chromatogram feature map is obtained. More specifically, the layers of the deep convolutional neural network model as the feature extractor are used to perform the following steps on the input data in the forward pass of the layers respectively: performing convolution processing on input data to obtain a convolution characteristic diagram; performing pooling processing based on a local feature matrix on the convolution feature map to obtain a pooled feature map; and performing nonlinear activation on the pooled feature map to obtain an activated feature map; the output of the last layer of the deep convolutional neural network serving as the feature extractor is the liquid chromatogram feature map, and the input of the first layer of the deep convolutional neural network serving as the feature extractor is the liquid chromatogram of the condensate water to be detected.
Fig. 4 is a flow chart of convolutional neural network coding in a measuring device of degassed hydrogen conductivity according to an embodiment of the application. As shown in fig. 4, in the convolutional neural network coding process, the convolutional neural network coding method includes: s210, performing convolution processing on input data to obtain a convolution characteristic diagram; s220, performing pooling processing based on a local feature matrix on the convolution feature map to obtain a pooled feature map; s230, carrying out nonlinear activation on the pooling feature map to obtain an activation feature map; the output of the last layer of the deep convolutional neural network serving as the feature extractor is the liquid chromatogram feature map, and the input of the first layer of the deep convolutional neural network serving as the feature extractor is the liquid chromatogram of the condensate water to be detected.
Specifically, during the operation of the degassed hydrogen conductivity measurement apparatus 300, the chromatogram feature enhancement module 340 is configured to pass the liquid chromatogram feature map through a parallel weight assignment module to obtain an enhanced liquid chromatogram feature map. It will be appreciated that in measuring the degassed hydrogen conductivity influence value of carbon dioxide, the content characteristics and characteristic characteristics of the carbonate ions therein, i.e. the hidden characteristics of the hidden distribution characteristics of the carbonate ions in spatial position and channel dimension, should be focused on while ignoring unwanted interference characteristics not related to the degassed hydrogen conductivity influence value measurement of carbon dioxide. Therefore, in the technical solution of the present application, in order to solve the problem that the target detection accuracy is low due to the edge blurring in the liquid chromatography feature map, a parallel weight assignment module is used to perform feature enhancement in the liquid chromatography feature map. Specifically, the liquid chromatogram characteristic diagram is processed by a parallel weight distribution module to obtain an enhanced liquid chromatogram characteristic diagram, so that effective characteristic representation can be enhanced, useless characteristic information can be suppressed, and the accuracy of subsequent classification can be improved. In particular, the parallel weight assignment module performs feature enhancement on the liquid chromatogram feature map by using a spatial attention module and a channel attention module respectively, wherein the extracted image features of the channel attention reflect the relevance and importance among feature channels, and the extracted image features of the spatial attention reflect the weight of feature difference of spatial dimensions, so as to suppress or enhance features of different spatial positions.
Fig. 5 is a block diagram of a chromatogram feature enhancement module in a measuring apparatus of degassed hydrogen conductivity according to an embodiment of the present application. As shown in fig. 5, the chromatogram feature enhancing module 340 includes: a spatial attention unit 341 configured to input the liquid chromatography feature map into a spatial attention module of the parallel weight assignment modules to obtain a spatial attention map; a channel attention unit 342 for inputting the liquid chromatography profile into a channel attention module of the parallel weight assignment module to obtain a channel attention map; and a fusion unit 343 for fusing the spatial attention map and the channel attention map to obtain the enhanced liquid chromatography profile. Wherein the spatial attention unit 341 comprises: the spatial perception subunit is used for carrying out convolutional coding on the liquid chromatogram characteristic diagram by using the convolutional layer of the spatial attention module to obtain a spatial attention matrix; a probabilistic subunit, configured to input the spatial attention moment array into a Softmax activation function of the spatial attention module to obtain a spatial attention score map; and a spatial attention applying subunit for calculating the spatial attention score map and the liquid chromatogram feature map by multiplying the position points to obtain the spatial attention map; the channel attention unit 342 includes: the global mean pooling subunit is used for calculating the global mean of each feature matrix of the liquid chromatogram feature map along the channel dimension to obtain a channel feature vector; a nonlinear activation subunit, configured to input the channel feature vector into a Softmax activation function to obtain a channel attention weight feature vector; and the channel attention applying subunit is used for respectively weighting each feature matrix of the liquid chromatogram feature map along the channel dimension by taking the feature value of each position in the channel attention weight feature vector as a weight so as to obtain the channel attention map.
Fig. 6 is a block diagram of a spatial attention cell in a degassed hydrogen conductivity measurement device according to an embodiment of the application. As shown in fig. 6, the spatial attention unit 341 includes: a spatial perception subunit 3411, configured to perform convolutional encoding on the liquid chromatogram feature map using convolutional layers of the spatial attention module to obtain a spatial attention matrix; a probabilistic subunit 3412, configured to input the spatial attention matrix into a Softmax activation function of the spatial attention module to obtain a spatial attention score map; and a spatial attention applying subunit 3413 configured to calculate a point-by-point multiplication of the spatial attention score map and the liquid chromatogram feature map to obtain the spatial attention map.
In particular, in the solution of the present application, the parallel weight assignment module derives from the liquid chromatography profile F 1 Obtaining a spatial attention diagram F by a spatial attention mechanism and a channel attention mechanism respectively 2 And channel attention map F 3 However, since the spatial attention mechanism and the channel attention mechanism respectively enhance feature extraction in different dimensions, this enables the spatial attention scheme F 2 And the channel attention map F 3 The overall feature distribution has a spatial position error in a high-dimensional feature space, which affects the fusion of the spatial attention diagrams F by means of point addition 2 And the channel attention map F 3 The fusion effect of (1).
In addition, the applicant of the present application considered the spatial attention map F 2 And the channel attention map F 3 Relative to theLiquid chromatogram characteristic graph F 1 Are homologous and thus have a correspondence so that the spatial attention map F can be applied 2 And the channel attention map F 3 The relative angle probability information representation correction is respectively carried out, and the representation is expressed as follows:
wherein
And &>
Characteristic values for the respective locations of the spatial attention map and the channel attention map, respectively, and +>
And &>
Is the mean of all the eigenvalues of the spatial attention map and the channel attention map, respectively, log representing the base-2 logarithmic function value. Here, the relative class angle probability information indicates that the correction is performed by the spatial attention map F 2 And the channel attention map F 3 Relative class angle probability information representation between to perform the space attention diagram F 2 And the channel attention map F 3 Geometric dilution of spatial position error of feature distribution in high-dimensional feature space, thereby performing F-shaped attention in the space 2 And the channel attention map F 3 Has a certain corresponding relation, based on the space attention diagram F 2 And the channel attention map F 3 The spatial attention map F is improved by performing implicit context correspondence correction of features by point-by-point regression of positions in comparison with the distribution constraint of the respective positions as a whole 2 And the channel attention map F 3 And the accuracy of decoding regression is improved by the fusion effect of point-and-add mode fusion. Thus, the real value of the degassed hydrogen conductivity of the condensed water can be accurately measured to ensure the safety of the equipment and the long-term economic benefit.
Fig. 7 is a block diagram of a fusion unit in a degassed hydrogen conductivity measurement apparatus according to an embodiment of the present application. As shown in fig. 7, the fusion unit 343 includes: a corrector subunit 3431, configured to perform feature correction on the spatial attention map and the channel attention map respectively to obtain a corrected spatial attention map and a corrected channel attention map according to the following formulas:
wherein
And &>
Is a characteristic value for the respective position of the spatial attention map and of the channel attention map, and->
And &>
Is the mean of all the characteristic values of the spatial attention map and the channel attention map, respectively, log represents the base-2 logarithmic function value; and a position-wise fusion subunit 3432 for calculating a position-wise sum between the corrected spatial attention map and the corrected channel attention map to obtain the enhanced liquid chromatography feature map.
Specifically, during the operation of the degassed hydrogen conductivity measurement device 300, the decoding module 350 is configured to pass the enhanced liquid chromatography characteristic map through a decoder to obtain a decoded value representing a degassed hydrogen conductivity influence value of carbon dioxide. And performing decoding regression by using the enhanced liquid chromatography characteristic diagram as a decoding characteristic diagram through a decoder to obtain a decoded value for representing the degassed hydrogen conductivity influence value of the carbon dioxide. That is, the content and characteristics of carbonate particles can be observed from the liquid chromatogram of the condensed water, and therefore, a decoded value for representing the influence of the degassed hydrogen conductivity of carbon dioxide is obtained based on the structure of the feature extractor + decoder. In one specific example of the present application, the enhanced liquid chromatography feature map is subjected to decoding regression using the decoder in the following formula to obtain a decoded value; wherein the formula is:
wherein X represents the enhanced liquid chromatography profile, Y is the decoded value, W is a weight matrix, and>
representing a matrix multiplication.
Specifically, during the operation of the device 300 for measuring the degassed hydrogen conductivity, the actual measurement value generation module 360 is configured to calculate a difference between the measured value of the degassed hydrogen conductivity and the decoded value to obtain an actual value of the degassed hydrogen conductivity measurement. In the technical scheme of the application, the difference value between the two is used for representing the degassing hydrogen conductivity measurement true value of the condensed water. Thus, the true value of the degassed hydrogen conductivity of the condensed water can be accurately measured to ensure the safety of the equipment and the long-term economic benefit.
In summary, the measuring apparatus 300 for degassed hydrogen conductivity according to the embodiment of the present application is illustrated, which obtains a degassed hydrogen conductivity measurement value of condensed water containing carbon dioxide influence through hydrogen conductivity meter measurement, and extracts implicit characteristic information of the content and characteristics of carbonate ions from a liquid chromatogram of the condensed water, so as to decode the degassed hydrogen conductivity influence value of carbon dioxide. Further, the difference between the two is used to represent the true value of the degassed hydrogen conductivity measurement of the condensed water. Thus, the real value of the degassed hydrogen conductivity of the condensed water can be accurately measured to ensure the safety of the equipment and the long-term economic benefit.
As described above, the measuring apparatus of degassed hydrogen conductivity according to the embodiments of the present application can be implemented in various terminal devices. In one example, the degassed hydrogen conductivity measurement apparatus 300 according to embodiments of the present application may be integrated into a terminal device as one software module and/or hardware module. For example, the measuring device 300 of the degassed hydrogen conductivity 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 degassed hydrogen conductivity measurement device 300 may also be one of many hardware modules of the terminal device.
Alternatively, in another example, the degassed hydrogen conductivity measurement apparatus 300 and the terminal device may be separate devices, and the degassed hydrogen conductivity measurement apparatus 300 may be connected to the terminal device through a wired and/or wireless network and transmit the mutual information according to an agreed data format.
Exemplary method
Fig. 8 is a flowchart of a method of measuring degassed hydrogen conductivity according to an embodiment of the application. As shown in fig. 8, the method for measuring the degassed hydrogen conductivity according to the embodiment of the present application includes the steps of: s110, obtaining a degassed hydrogen conductivity measurement value of the condensate to be detected through a common hydrogen conductivity meter; s120, acquiring a liquid chromatogram of the condensate to be detected; s130, enabling the liquid chromatogram of the condensed water to be detected to pass through a deep convolution neural network model serving as a feature extractor to obtain a liquid chromatogram feature map; s140, the liquid chromatogram characteristic diagram is processed by a parallel weight distribution module to obtain an enhanced liquid chromatogram characteristic diagram; s150, enabling the enhanced liquid chromatography characteristic diagram to pass through a decoder to obtain a decoding value used for representing the degassed hydrogen conductivity influence value of the carbon dioxide; and S160, calculating the difference value between the measured value of the degassed hydrogen conductivity and the decoded value to obtain a true value of the degassed hydrogen conductivity measurement.
In one example, in the method for measuring a conductivity of degassed hydrogen, the step S130 includes: performing, in a layer forward pass, input data using the layers of the deep convolutional neural network model as a feature extractor: performing convolution processing on input data to obtain a convolution characteristic diagram; performing pooling processing based on a local feature matrix on the convolution feature map to obtain a pooled feature map; and performing nonlinear activation on the pooled feature map to obtain an activated feature map; the output of the last layer of the deep convolutional neural network serving as the feature extractor is the liquid chromatogram feature map, and the input of the first layer of the deep convolutional neural network serving as the feature extractor is the liquid chromatogram of the condensate to be detected.
In one example, in the method for measuring the conductivity of the degassed hydrogen, the step S140 includes: inputting the liquid chromatography profile into a spatial attention module of the parallel weight assignment module to obtain a spatial attention map; inputting the liquid chromatography profile into a channel attention module of the parallel weight assignment module to obtain a channel attention map; and fusing the spatial attention map and the channel attention map to obtain the enhanced liquid chromatography signature. Wherein the inputting the liquid chromatography profile into a spatial attention module of the parallel weight assignment module to obtain a spatial attention map comprises: convolution encoding the liquid chromatography signature using convolution layers of the spatial attention module to obtain a spatial attention matrix; inputting the spatial attention moment array into a Softmax activation function of the spatial attention module to obtain a spatial attention score map; and calculating the spatial attention score map and the liquid chromatogram feature map by multiplying the position points to obtain the spatial attention map. The inputting the liquid chromatography profile into a channel attention module of the parallel weight assignment module to obtain a channel attention map, comprising: calculating a global mean value of each feature matrix of the liquid chromatogram feature map along the channel dimension to obtain a channel feature vector; inputting the channel feature vector into a Softmax activation function to obtain a channel attention weight feature vector; and respectively weighting each feature matrix of the liquid chromatogram feature map along the channel dimension by taking the feature value of each position in the channel attention weight feature vector as a weight so as to obtain the channel attention map. More specifically, the fusing the spatial attention map and the channel attention map to obtain the enhanced liquid chromatography profile comprises: respectively performing characteristic correction on the spatial attention diagram and the channel attention diagram to obtain a corrected spatial attention diagram and a corrected channel attention diagram according to the following formulas:
wherein
And &>
Is a characteristic value for the respective position of the spatial attention map and of the channel attention map, and->
And &>
Is the mean of all the characteristic values of the spatial attention map and the channel attention map, respectively, log represents the base-2 logarithmic function value; and calculating a position-wise sum between the corrected spatial attention map and the corrected channel attention map to obtain the enhanced liquid chromatography feature map.
In one example, in the method for measuring the dehydrogenation hydrogen conductivity, the step S150 includes: performing decoding regression on the enhanced liquid chromatography characteristic map by using the decoder according to the following formula to obtain a decoded value; wherein the formula is:
wherein X represents the enhanced liquid chromatography profile, Y is the decoded value, W is a weight matrix, and W is a->
Representing a matrix multiplication.
In summary, the method for measuring the degassed hydrogen conductivity according to the embodiment of the present application is illustrated, which obtains a degassed hydrogen conductivity measurement value of the condensed water containing the influence of carbon dioxide through hydrogen conductivity meter measurement, and extracts implicit characteristic information of the content and characteristics of carbonate ions from the liquid chromatogram of the condensed water, so as to decode the obtained degassed hydrogen conductivity influence value of carbon dioxide. Further, the difference between the two is used to represent the true value of the degassed hydrogen conductivity measurement of the condensed water. Thus, the real value of the degassed hydrogen conductivity of the condensed water can be accurately measured to ensure the safety of the equipment and the long-term economic benefit.
Claims (10)
1. A measuring device for degassed hydrogen conductivity, comprising:
the common measurement module is used for obtaining a degassed hydrogen conductivity measurement value of the condensate water to be detected through a common hydrogen conductivity meter;
the liquid chromatogram acquisition module is used for acquiring a liquid chromatogram of the to-be-detected condensed water;
the liquid chromatogram characteristic extraction module is used for enabling the liquid chromatogram of the condensate water to be detected to pass through a deep convolution neural network model serving as a characteristic extractor to obtain a liquid chromatogram characteristic diagram;
the chromatogram characteristic enhancement module is used for enabling the liquid chromatogram characteristic graph to pass through the parallel weight distribution module to obtain an enhanced liquid chromatogram characteristic graph;
the decoding module is used for enabling the enhanced liquid chromatography characteristic diagram to pass through a decoder to obtain a decoded value used for representing a degassed hydrogen conductivity influence value of the carbon dioxide; and
and the measurement true value generation module is used for calculating the difference value between the degassing hydrogen conductivity measurement value and the decoding value so as to obtain the degassing hydrogen conductivity measurement true value.
2. The degassed hydrogen conductivity measurement apparatus according to claim 1, wherein the liquid chromatogram feature extraction module is further configured to: performing, in a layer forward pass, input data using the layers of the deep convolutional neural network model as a feature extractor:
performing convolution processing on input data to obtain a convolution characteristic diagram;
performing pooling processing based on a local feature matrix on the convolution feature map to obtain a pooled feature map; and
performing nonlinear activation on the pooled feature map to obtain an activated feature map;
the output of the last layer of the deep convolutional neural network serving as the feature extractor is the liquid chromatogram feature map, and the input of the first layer of the deep convolutional neural network serving as the feature extractor is the liquid chromatogram of the condensate water to be detected.
3. The degassed hydrogen conductivity measurement device according to claim 2, wherein said chromatogram feature enhancement module comprises:
a spatial attention unit for inputting the liquid chromatography profile into a spatial attention module of the parallel weight assignment module to obtain a spatial attention map;
a channel attention unit for inputting the liquid chromatography profile into a channel attention module of the parallel weight assignment module to obtain a channel attention map; and
a fusion unit for fusing the spatial attention map and the channel attention map to obtain the enhanced liquid chromatography profile.
4. The degassed hydrogen conductivity measurement device according to claim 3, wherein the spatial attention unit comprises:
a spatial perception subunit, configured to perform convolutional encoding on the liquid chromatography feature map using convolutional layers of the spatial attention module to obtain a spatial attention matrix;
a probabilistic subunit, configured to input the spatial attention moment array into a Softmax activation function of the spatial attention module to obtain a spatial attention score map; and
a spatial attention applying subunit for calculating the spatial attention score map and the liquid chromatography feature map by multiplying the position points to obtain the spatial attention map.
5. The degassed hydrogen conductivity measurement device according to claim 4, wherein said channel attention unit comprises:
the global mean pooling subunit is used for calculating the global mean of each feature matrix of the liquid chromatogram feature map along the channel dimension to obtain a channel feature vector;
a nonlinear activation subunit, configured to input the channel feature vector into a Softmax activation function to obtain a channel attention weight feature vector; and
and the channel attention applying subunit is used for respectively weighting each feature matrix of the liquid chromatogram feature map along the channel dimension by taking the feature value of each position in the channel attention weight feature vector as a weight so as to obtain the channel attention map.
6. The degassed hydrogen conductivity measurement device according to claim 5, wherein the fusion unit comprises:
a corrector subunit, configured to perform feature correction on the spatial attention diagram and the channel attention diagram respectively to obtain a corrected spatial attention diagram and a corrected channel attention diagram, with the following formulas:
wherein
And &>
Is a characteristic value for the respective position of the spatial attention map and of the channel attention map, and->
And &>
Is the mean of all the characteristic values of the spatial attention map and the channel attention map, respectively, log represents the base-2 logarithmic function value; and
a position-wise fusion subunit for calculating a position-wise sum between the corrected spatial attention map and the corrected channel attention map to obtain the enhanced liquid chromatography feature map.
7. The degassed hydrogen conductivity measurement apparatus according to claim 6, wherein said decoding module is further configured to: performing decoding regression on the enhanced liquid chromatography characteristic map by using the decoder according to the following formula to obtain a decoded value;
wherein the formula is:
wherein X represents the enhanced liquid chromatography profile, Y is the decoded value, W is a weight matrix, and>
representing a matrix multiplication.
8. A method for measuring the electrical conductivity of degassed hydrogen, comprising:
obtaining a measured value of the degassed hydrogen conductivity of the condensate water to be detected through a common hydrogen conductivity meter;
acquiring a liquid chromatogram of the condensate to be detected;
the liquid chromatogram of the condensate to be detected is processed by a deep convolution neural network model as a feature extractor to obtain a liquid chromatogram feature map;
passing the liquid chromatography profile through a parallel weight assignment module to obtain an enhanced liquid chromatography profile;
passing the enhanced liquid chromatography signature through a decoder to obtain a decoded value representing a degassed hydrogen conductivity influence value of carbon dioxide; and
and calculating the difference value between the measured value of the degassed hydrogen conductivity and the decoded value to obtain a true value of the degassed hydrogen conductivity measurement.
9. The method for measuring degassed hydrogen conductivity according to claim 8, wherein the passing the liquid chromatogram of the condensate to be detected through a deep convolutional neural network model as a feature extractor to obtain a liquid chromatogram feature map comprises: performing, in a layer forward pass, input data using the layers of the deep convolutional neural network model as a feature extractor:
performing convolution processing on input data to obtain a convolution characteristic diagram;
performing pooling processing based on a local feature matrix on the convolution feature map to obtain a pooled feature map; and
performing nonlinear activation on the pooled feature map to obtain an activated feature map;
the output of the last layer of the deep convolutional neural network serving as the feature extractor is the liquid chromatogram feature map, and the input of the first layer of the deep convolutional neural network serving as the feature extractor is the liquid chromatogram of the condensate water to be detected.
10. The method for measuring degassed hydrogen conductivity according to claim 9, wherein said passing said enhanced liquid chromatography signature through a decoder to obtain decoded values representing degassed hydrogen conductivity influence values of carbon dioxide comprises: performing decoding regression on the enhanced liquid chromatography characteristic map by using the decoder according to the following formula to obtain a decoded value;
wherein the formula is:
wherein X represents the enhanced liquid chromatography profile,y is the decoded value, W is a weight matrix, and>
representing a matrix multiplication. />
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211388392.4A CN115902004B (en) | 2022-11-08 | 2022-11-08 | A kind of measuring device and measuring method of conductivity of degassed hydrogen |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211388392.4A CN115902004B (en) | 2022-11-08 | 2022-11-08 | A kind of measuring device and measuring method of conductivity of degassed hydrogen |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115902004A true CN115902004A (en) | 2023-04-04 |
| CN115902004B CN115902004B (en) | 2023-08-18 |
Family
ID=86477714
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211388392.4A Active CN115902004B (en) | 2022-11-08 | 2022-11-08 | A kind of measuring device and measuring method of conductivity of degassed hydrogen |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115902004B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116106457A (en) * | 2023-04-13 | 2023-05-12 | 天津海河标测技术检测有限公司 | Air sampling and detecting integrated device |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH01277748A (en) * | 1988-04-28 | 1989-11-08 | Hitachi Ltd | Water quality monitor method in steam power plant |
| US5252491A (en) * | 1990-03-26 | 1993-10-12 | The Babcock & Wilcox Company | Aqueous carbon dioxide monitor |
| US5275957A (en) * | 1984-01-10 | 1994-01-04 | Anatel Corporation | Instrument and method for measurement of the organic carbon content of water |
| JPH11258191A (en) * | 1998-03-09 | 1999-09-24 | Japan Organo Co Ltd | Method and apparatus for measurement of concentration of carbonic acid |
| CN1479094A (en) * | 2002-08-27 | 2004-03-03 | 上海电力学院 | Method and device for eliminating carbon dioxide influence in hydrogen electric conductivity measurement |
| CN101672813A (en) * | 2008-09-10 | 2010-03-17 | 华东电力试验研究院有限公司 | Method for detecting content of traces of carbon dioxide in water |
| JP2012013621A (en) * | 2010-07-02 | 2012-01-19 | Tokyo Electric Power Co Inc:The | Quantitative determination method for dissolved carbonic acid component |
| US20120178175A1 (en) * | 2011-01-12 | 2012-07-12 | Jay Clifford Crosman | CWB conductivity monitor |
| CN107064232A (en) * | 2017-01-23 | 2017-08-18 | 北京华科仪科技股份有限公司 | The measuring method and device of a kind of power plant's water degasification hydrogen conductivity |
| CN109030705A (en) * | 2018-11-01 | 2018-12-18 | 中科谱研（北京）科技有限公司 | A kind of method of the ion-chromatographic determination containing carbonic acid (hydrogen) root in carbonic acid (hydrogen) salt injection |
| CN110779961A (en) * | 2019-11-06 | 2020-02-11 | 中国科学院生态环境研究中心 | A kind of bicarbonate detection device and its using method and dialysis machine |
| CN111443160A (en) * | 2020-01-23 | 2020-07-24 | 华东理工大学 | Gas-sensitive-gas chromatography information fusion and electronic nose instrument on-line analysis method |
| CN111638249A (en) * | 2020-05-31 | 2020-09-08 | 天津大学 | Water content measuring method based on deep learning and application of water content measuring method in oil well exploitation |
| CN113030002A (en) * | 2021-03-15 | 2021-06-25 | 中国大唐集团科学技术研究院有限公司华中电力试验研究院 | Water vapor quality abnormity determination method and monitoring equipment based on infrared spectroscopy |
| CN113671057A (en) * | 2020-05-15 | 2021-11-19 | 中国科学院理化技术研究所 | Method for detecting aerobic biodegradation performance of high polymer material in marine environment |
| CN113740386A (en) * | 2021-08-20 | 2021-12-03 | 海南核电有限公司 | Measuring device and measuring method for accurately measuring hydrogen conductivity of water |
| CN114910519A (en) * | 2022-04-29 | 2022-08-16 | 华能国际电力股份有限公司 | Method and system for detecting conductivity of degassed hydrogen |
| CN114910518A (en) * | 2022-04-29 | 2022-08-16 | 浙江西热利华智能传感技术有限公司 | Detection method and detection system of dissolved carbon dioxide content in water vapor |
-
2022
- 2022-11-08 CN CN202211388392.4A patent/CN115902004B/en active Active
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5275957A (en) * | 1984-01-10 | 1994-01-04 | Anatel Corporation | Instrument and method for measurement of the organic carbon content of water |
| JPH01277748A (en) * | 1988-04-28 | 1989-11-08 | Hitachi Ltd | Water quality monitor method in steam power plant |
| US5252491A (en) * | 1990-03-26 | 1993-10-12 | The Babcock & Wilcox Company | Aqueous carbon dioxide monitor |
| JPH11258191A (en) * | 1998-03-09 | 1999-09-24 | Japan Organo Co Ltd | Method and apparatus for measurement of concentration of carbonic acid |
| CN1479094A (en) * | 2002-08-27 | 2004-03-03 | 上海电力学院 | Method and device for eliminating carbon dioxide influence in hydrogen electric conductivity measurement |
| CN101672813A (en) * | 2008-09-10 | 2010-03-17 | 华东电力试验研究院有限公司 | Method for detecting content of traces of carbon dioxide in water |
| JP2012013621A (en) * | 2010-07-02 | 2012-01-19 | Tokyo Electric Power Co Inc:The | Quantitative determination method for dissolved carbonic acid component |
| US20120178175A1 (en) * | 2011-01-12 | 2012-07-12 | Jay Clifford Crosman | CWB conductivity monitor |
| CN107064232A (en) * | 2017-01-23 | 2017-08-18 | 北京华科仪科技股份有限公司 | The measuring method and device of a kind of power plant's water degasification hydrogen conductivity |
| CN109030705A (en) * | 2018-11-01 | 2018-12-18 | 中科谱研（北京）科技有限公司 | A kind of method of the ion-chromatographic determination containing carbonic acid (hydrogen) root in carbonic acid (hydrogen) salt injection |
| CN110779961A (en) * | 2019-11-06 | 2020-02-11 | 中国科学院生态环境研究中心 | A kind of bicarbonate detection device and its using method and dialysis machine |
| CN111443160A (en) * | 2020-01-23 | 2020-07-24 | 华东理工大学 | Gas-sensitive-gas chromatography information fusion and electronic nose instrument on-line analysis method |
| CN113671057A (en) * | 2020-05-15 | 2021-11-19 | 中国科学院理化技术研究所 | Method for detecting aerobic biodegradation performance of high polymer material in marine environment |
| CN111638249A (en) * | 2020-05-31 | 2020-09-08 | 天津大学 | Water content measuring method based on deep learning and application of water content measuring method in oil well exploitation |
| CN113030002A (en) * | 2021-03-15 | 2021-06-25 | 中国大唐集团科学技术研究院有限公司华中电力试验研究院 | Water vapor quality abnormity determination method and monitoring equipment based on infrared spectroscopy |
| CN113740386A (en) * | 2021-08-20 | 2021-12-03 | 海南核电有限公司 | Measuring device and measuring method for accurately measuring hydrogen conductivity of water |
| CN114910519A (en) * | 2022-04-29 | 2022-08-16 | 华能国际电力股份有限公司 | Method and system for detecting conductivity of degassed hydrogen |
| CN114910518A (en) * | 2022-04-29 | 2022-08-16 | 浙江西热利华智能传感技术有限公司 | Detection method and detection system of dissolved carbon dioxide content in water vapor |
Non-Patent Citations (6)
| Title |
|---|
| MD. OMER FARUQUE等: "A comparative analysis to forecast carbon dioxide emissions", ENERGY REPORTS, vol. 8, pages 8046 - 8060 * |
| 王军等: "基于CNN-LSTM模型的黄河水质预测研究", 人民黄河, vol. 43, no. 5, pages 96 - 99 * |
| 王志勃等: "基于CNN-LSTM的白马湖水质预测模型研究", 电脑知识与技术, vol. 18, no. 26, pages 11 - 13 * |
| 秦宇等: "基于CNN-LSTM模型及小样本数据的水库二氧化碳通量预测", 重庆交通大学学报(自然科学版), vol. 41, no. 6, pages 119 - 125 * |
| 蔡青松等: "离子色谱法测定油田有色地层水中的碳酸根和碳酸氢根", 分析测试技术与仪器, no. 3, pages 165 - 169 * |
| 谢磊等: "离子色谱法测定自制直饮纯净水的6种阴离子", 广州化工, vol. 44, no. 10, pages 144 - 146 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116106457A (en) * | 2023-04-13 | 2023-05-12 | 天津海河标测技术检测有限公司 | Air sampling and detecting integrated device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115902004B (en) | 2023-08-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114998673B (en) | Dam defect time sequence image description method based on local self-attention mechanism | |
| CN115324843B (en) | Wind generating set fault diagnosis system and method based on monitoring data | |
| WO2023226228A1 (en) | Intelligent toxic and harmful gas alarm system for preparation of electronic grade hydrogen peroxide | |
| CN115235612A (en) | Intelligent fault diagnosis system and method for servo motor | |
| CN115754107B (en) | Automatic sampling analysis system and method for preparation of lithium hexafluorophosphate | |
| CN115600140A (en) | Fan variable pitch system fault identification method and system based on multi-source data fusion | |
| CN115754108B (en) | Acidity determination system and method for electronic grade hexafluorobutadiene | |
| CN115018012A (en) | A method and system for anomaly detection of Internet of Things time series under high-dimensional characteristics | |
| CN115984745A (en) | Moisture Control Method for Black Garlic Fermentation | |
| CN115902004A (en) | Measuring device and measuring method for conductivity of degassed hydrogen | |
| Yu et al. | Unlocking the capabilities of large vision-language models for generalizable and explainable deepfake detection | |
| CN115471216A (en) | Data management method of smart laboratory management platform | |
| CN117612025B (en) | Roof recognition method of remote sensing images based on diffusion model | |
| CN118822048A (en) | A sub-seasonal extreme precipitation prediction method based on deep fusion of precursor factors | |
| CN118644663A (en) | An infrared sea surface target detection method based on global-local fusion attention | |
| CN117176270A (en) | Indoor antenna with signal monitoring function | |
| CN116503314A (en) | Quality inspection system and method for door manufacturing | |
| CN119415848B (en) | Missing modal data processing method based on mutual information constraint and countermeasure learning | |
| Karimi et al. | Dsv-lfs: Unifying llm-driven semantic cues with visual features for robust few-shot segmentation | |
| CN116467485B (en) | Video image retrieval construction system and method thereof | |
| CN115049851B (en) | Target detection method, device and equipment terminal based on YOLOv5 network | |
| CN114357232B (en) | Processing methods, systems, devices, and storage media for extracting ship track features | |
| CN117876803A (en) | A method and device for detecting small infrared targets based on orthogonal subspace projection | |
| CN115723280A (en) | Production equipment of polyimide film with adjustable thickness | |
| CN117392432A (en) | Agricultural product detection methods, devices, electronic equipment and computer-readable storage media |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A measuring device and method for the conductivity of degassed hydrogen Effective date of registration: 20231115 Granted publication date: 20230818 Pledgee: Qilu bank Limited by Share Ltd Ji'nan branch Wangsheren Pledgor: Zhongketken (Shandong) Intelligent Technology Co.,Ltd. Registration number: Y2023980065873 |