CN117556245B - Method for detecting filtered impurities in tetramethylammonium hydroxide production - Google Patents
Method for detecting filtered impurities in tetramethylammonium hydroxide production Download PDFInfo
- Publication number
- CN117556245B CN117556245B CN202410009120.1A CN202410009120A CN117556245B CN 117556245 B CN117556245 B CN 117556245B CN 202410009120 A CN202410009120 A CN 202410009120A CN 117556245 B CN117556245 B CN 117556245B
- Authority
- CN
- China
- Prior art keywords
- spectrum data
- sequence
- independent component
- independent
- cluster
- 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
- WGTYBPLFGIVFAS-UHFFFAOYSA-M tetramethylammonium hydroxide Chemical compound [OH-].C[N+](C)(C)C WGTYBPLFGIVFAS-UHFFFAOYSA-M 0.000 title claims abstract description 154
- 239000012535 impurity Substances 0.000 title claims abstract description 70
- 238000000034 method Methods 0.000 title claims abstract description 69
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 22
- 238000001228 spectrum Methods 0.000 claims abstract description 235
- 230000003595 spectral effect Effects 0.000 claims description 47
- 238000012880 independent component analysis Methods 0.000 claims description 25
- 238000010521 absorption reaction Methods 0.000 claims description 17
- 238000001514 detection method Methods 0.000 claims description 9
- 238000005457 optimization Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 3
- 239000000126 substance Substances 0.000 description 8
- 230000000694 effects Effects 0.000 description 5
- 239000006185 dispersion Substances 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000001259 photo etching Methods 0.000 description 1
- 238000000206 photolithography Methods 0.000 description 1
- 229920002120 photoresistant polymer Polymers 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2134—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on separation criteria, e.g. independent component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Abstract
The invention relates to the technical field of spectrum data processing, in particular to a method for detecting filtered impurities in tetramethylammonium hydroxide production, which comprises the following steps: collecting spectrum data of the tetramethylammonium hydroxide solution to obtain independent components of the spectrum data, obtaining ordered independent components of the spectrum data according to the independent components, matching wavelength pairs when DTW matching is carried out according to the ordered independent components of the spectrum data, the independent components of the spectrum data in the class clusters and the independent components of different spectrum data in the class clusters, obtaining the preference degree of splitting the class clusters into two class clusters when iterative self-organizing clustering is carried out, carrying out iterative self-organizing clustering on all the spectrum data according to the preference degree and preset parameters, and obtaining main impurities of the tetramethylammonium hydroxide solution corresponding to the final class clusters. The invention determines the final splitting mode by utilizing the discrete degree among the independent components, so that the splitting results can better embody the information of different clusters and reduce the misjudgment probability of impurities.
Description
Technical Field
The invention relates to the technical field of spectrum data processing, in particular to a method for detecting filtered impurities in tetramethylammonium hydroxide production.
Background
In the photolithography process, tetramethyl ammonium hydroxide (TMAH) is a commonly used developer, and is mainly used for developing photoresist, when the tetramethyl ammonium hydroxide solution contains impurities, the developing effect is affected, for example, the impurities may cause local chemical reaction differences in the developing solution, the surface after development is uneven or cannot be accurately developed, and by performing cluster analysis on spectral data of the tetramethyl ammonium hydroxide solution, the impurities can be rapidly identified and classified from a large amount of tetramethyl ammonium hydroxide solution.
When the existing method utilizes ISODATA iterative self-organizing clustering, when judging that one class cluster needs to be split, the method only selects the sub-class clustering center point according to the mean value of the corresponding class cluster and the standard deviation of part of dimensions, and the selected class cluster center point may not represent the condition of high-density points, so that the same class cluster is divided or different class clusters are divided into the same class cluster, and further the impurity detection of the tetramethylammonium hydroxide solution is inaccurate.
Disclosure of Invention
In order to solve the problems, the invention provides a method for detecting the filtered impurities in the production of tetramethylammonium hydroxide.
The invention relates to a method for detecting filtered impurities in tetramethylammonium hydroxide production, which adopts the following technical scheme:
An embodiment of the invention provides a method for detecting filtered impurities in tetramethyl ammonium hydroxide production, which comprises the following steps:
collecting spectrum data of a plurality of tetramethyl ammonium hydroxide solutions;
acquiring a plurality of independent components of each spectrum data, acquiring a plurality of characteristic wavelengths of each independent component of each spectrum data according to the wavelength in the independent components and the absorption rate corresponding to the wavelength, and acquiring a plurality of independent component sequences and ordered independent components of other spectrum data under each spectrum data according to a plurality of characteristic wavelengths of each independent component of the spectrum data;
when all spectrum data are obtained for iterative self-organizing clustering, a plurality of clusters are clustered for the first time, wavelength pairs are matched according to ordered independent components of each spectrum data, independent components of spectrum data in each cluster and independent components of different spectrum data in each cluster when DTW matching is carried out, so that the similarity degree between the independent components of different spectrum data in each cluster is obtained, the discrete degree of the independent components of the spectrum data in each cluster in the corresponding independent component sequence is obtained according to the similarity degree of the independent components of the spectrum data in each cluster in the corresponding independent component sequence, and the preference degree of splitting each cluster into two clusters when iterative self-organizing clustering is carried out is obtained;
According to the optimization degree and preset parameters of each class cluster divided into two class clusters when iterative self-organizing clustering is carried out, iterative self-organizing clustering is carried out on all spectrum data to obtain a plurality of final class clusters, and impurity detection is carried out on part of the tetramethylammonium hydroxide solution corresponding to the final class clusters to obtain main impurities of the tetramethylammonium hydroxide solution corresponding to each final class cluster.
Further, the obtaining a plurality of characteristic wavelengths of each independent component of each spectrum data according to the wavelength and the absorption rate corresponding to the wavelength in the independent component comprises the following specific steps:
for any one of the individual components of any one of the spectral data, ifAnd->The corresponding independent component is in the form of +.>The wavelength is the center, the neighborhood radius is +.>The sum of the absorptances corresponding to all wavelengths in the wavelength range of (2) is greater than 1, and +.>The individual wavelengths are taken as the independent componentsA characteristic wavelength of the quantity;
wherein the method comprises the steps ofRepresenting the +.>Absorption corresponding to the individual wavelength, < >>For a preset second value of the number,representing the independent component as +.>The wavelength is the center, the neighborhood radius is +.>In the wavelength range->Absorption corresponding to the individual wavelength, < >>。
Further, according to the characteristic wavelengths of each independent component of the spectrum data, a plurality of independent component sequences and ordered independent components of other spectrum data under each spectrum data are obtained, and the method comprises the following specific steps:
Recording any one spectrum data as target spectrum data, artificially numbering all independent components of the target spectrum data to obtain a plurality of independent components with an arrangement sequence, and arranging all characteristic wavelengths of a first independent component of the target spectrum data in a sequence from small to large to obtain a characteristic wavelength sequence of the first independent component of the target spectrum data;
the characteristic wavelength sequence of the first independent component of the target spectrum data is recorded as a first sequence, and the characteristic wavelength sequence with the highest similarity with the first sequence is obtained from the characteristic wavelength sequences of all independent components of any spectrum data except the target spectrum data and recorded as a similar characteristic wavelength sequence of the first sequence;
in the characteristic wavelength sequences of all independent components of each spectrum data except the target spectrum data, acquiring a plurality of similar characteristic wavelength sequences of a first sequence, arranging the independent components corresponding to the first sequence and the independent components corresponding to the plurality of similar characteristic wavelength sequences of the first sequence from large to small according to the similarity of the characteristic wavelength sequences to obtain a long sequence, and marking the long sequence as an independent component sequence corresponding to the first independent component of the target spectrum data, wherein the independent component corresponding to the first sequence is arranged at the first position of the independent component sequence, and meanwhile, the independent components corresponding to the plurality of similar characteristic wavelength sequences of the first sequence are arranged as the first independent component in the spectrum data to which the independent components corresponding to the first sequence belong;
The characteristic wavelength sequence of the second independent component of the target spectrum data is recorded as a second sequence, and the characteristic wavelength sequence with the highest similarity with the second sequence is obtained from the characteristic wavelength sequences of all independent components of any spectrum data except the target spectrum data and recorded as a similar characteristic wavelength sequence of the second sequence;
in the characteristic wavelength sequences of all independent components of each spectrum data except the target spectrum data, acquiring a plurality of similar characteristic wavelength sequences of a second sequence, arranging the independent components corresponding to the second sequence and the independent components corresponding to the similar characteristic wavelength sequences of the second sequence from large to small according to the similarity of the characteristic wavelength sequences to obtain a long sequence, and marking the long sequence as an independent component sequence corresponding to the second independent component of the target spectrum data, wherein the independent component corresponding to the second sequence is arranged at a first position of the independent component sequence, and meanwhile, the independent components corresponding to the similar characteristic wavelength sequences of the second sequence are arranged as a second independent component in the spectrum data;
and the like, obtaining an independent component sequence corresponding to each independent component of the target spectrum data and ordered independent components of other spectrum data under the target spectrum data.
Further, the specific acquisition method of the similarity is as follows:
and obtaining the similarity between the characteristic wavelength sequence and the first sequence by using a DTW algorithm.
Further, the matching wavelength pairs when DTW matching is performed according to the ordered independent components of each spectrum data, the independent components of each spectrum data in each class cluster, and the independent components of different spectrum data in each class cluster, so as to obtain the similarity degree between the independent components of different spectrum data in each class cluster, which comprises the following specific steps:
any one of the class clusters is marked as a target class cluster;
in the method, in the process of the invention,for the p-th independent component of the i-th spectral data in the target class cluster,/for the i-th spectral data in the target class cluster>For the p-th independent component of the j-th spectral data in the target class cluster,/for>The pearson correlation coefficient for the p-th independent component of the ith spectral data in the target class cluster and the p-th independent component of the jth spectral data in the target class cluster,/>When DTW matching is carried out on the p independent component of the ith spectrum data in the target class cluster and the p independent component of the jth spectrum data in the target class cluster, the number of the matched wavelength pairs is matched; />For the wavelength of the p-th independent component belonging to the i-th spectral data in the k-th matched wavelength pair, is- >For the wavelength of the p-th independent component belonging to the j-th spectral data in the k-th matched wavelength pair, is>Is the +.>First->Weight coefficient of individual components, +.>Is the +.>First->The weighting coefficients of the individual components, where +.>The sum of the individual spectral data->Weight coefficient and +.>First->The weight coefficients of the individual components can be obtained by ICA individual component analysis, < >>As an exponential function based on natural constants, < +.>To take absolute value, +.>The similarity degree between the p independent component of the ith spectrum data in the target class cluster and the p independent component of the jth spectrum data in the target class cluster is obtained.
Further, according to the degree of similarity between the independent components of different spectrum data in each class cluster, the discrete degree of the independent components of the spectrum data in each class cluster in the corresponding independent component sequence is obtained, which comprises the following specific steps:
in the method, in the process of the invention,for the number of spectral data in the target class cluster, < >>For the similarity degree of the p-th independent component of the ith spectrum data in the target class cluster and the p-th independent component of the jth spectrum data in the target class cluster, < > >To avoid hyper-parameters with denominator 0, < ->The p-th independent component of the spectrum data in the target class cluster is the discrete degree of the corresponding p-th independent component sequence.
Further, according to the discrete degree of the independent component of the spectrum data in each class cluster in the corresponding independent component sequence, the optimization degree of splitting each class cluster into two class clusters when iterative self-organizing clustering is performed is obtained, and the method comprises the following specific steps:
when any one class cluster is subjected to iterative self-organizing clustering, the two class clusters divided into the class cluster are respectively marked as a first sub-class cluster and a second sub-class cluster;
in the method, in the process of the invention,for the degree of discretization of the p-th independent component of the spectral data in the first sub-cluster in the sequence of the corresponding p-th independent component,>for the degree of discretization of the p-th independent component of the spectral data in the second sub-cluster in the sequence of the corresponding p-th independent component,>for the distance between the centers of the first sub-cluster and the second sub-cluster, < >>For the number of independent components per spectral data, +.>Preference for splitting the class cluster into a first sub-cluster and a second sub-cluster, +.>Is a linear normalization function.
Further, the method for collecting the spectrum data of a plurality of tetramethyl ammonium hydroxide solutions comprises the following specific steps:
A first numerical value is preset and is recorded as N, and a multispectral instrument is utilized to respectively obtain the spectrum data of N tetramethyl ammonium hydroxide solutions.
Further, the specific method for acquiring the independent components of each spectrum data is as follows:
and (3) performing ICA independent component analysis on each spectrum data to obtain a plurality of independent components of each spectrum data.
Further, the method for detecting the impurities of the part of the tetramethylammonium hydroxide solution corresponding to the final cluster to obtain the main impurities of the tetramethylammonium hydroxide solution corresponding to each final cluster comprises the following specific steps:
randomly sampling and selecting N1 tetramethyl ammonium hydroxide solutions corresponding to the spectrum data in any one final cluster, respectively detecting impurities of the selected N1 tetramethyl ammonium hydroxide solutions, taking the impurity with the highest impurity content as the main impurity of the tetramethyl ammonium hydroxide solution corresponding to the final cluster, wherein N1 is a preset third value.
The technical scheme of the invention has the beneficial effects that: according to the invention, by collecting the spectrum data of the tetramethyl ammonium hydroxide solution, analyzing the similarity degree of independent components of the spectrum data and analyzing the discrete degree of independent component sequences, the optimization degree of each cluster divided into two clusters during iterative self-organizing clustering is obtained, so that the main impurity of the tetramethyl ammonium hydroxide solution is obtained, and the detection of the impurity is completed; aiming at the problem that the misjudgment probability is higher due to the traditional cluster splitting result in the iterative self-organizing clustering process, ICA decomposition is carried out on the spectrum data of different tetramethyl ammonium hydroxide solutions, the similarity degree of the independent components is obtained by utilizing the correlation coefficient between the independent components and the ratio between the matched wavelength pairs when the independent components are subjected to DTW matching, and then the discrete degree of the independent component sequences is obtained according to the similarity degree between the independent components of different spectrum data in each cluster, so that the final splitting mode is determined, the information of different clusters can be better reflected by the splitting result, and the misjudgment probability of impurities is reduced.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flow chart of steps of a method for detecting impurities in tetramethyl ammonium hydroxide production according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to specific implementation, structure, characteristics and effects of a method for detecting the filtered impurities in the production of tetramethylammonium hydroxide according to the invention by combining the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the method for detecting the filtered impurities in the production of tetramethylammonium hydroxide provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for detecting impurities in tetramethyl ammonium hydroxide production according to an embodiment of the invention is shown, the method comprises the following steps:
and S001, collecting spectrum data of a plurality of tetramethyl ammonium hydroxide solutions.
It should be noted that the purpose of this embodiment is to detect impurities contained in the tetramethylammonium hydroxide solution, and several spectroscopic data are first collected before starting the analysis.
Specifically, a first numerical value is preset and is denoted as N, in this embodiment, n=30 is used to describe, and a multispectral instrument is used to obtain spectral data of N tetramethylammonium hydroxide solutions respectively, where the spectral data is two-dimensional data, the horizontal axes of the spectral data are different wavelengths, the vertical axes of the spectral data are absorption rates corresponding to the different wavelengths, and the minimum difference of the different wavelengths is 1nm.
To this end, several spectral data are obtained.
And step S002, ICA independent component analysis is carried out on the spectrum data to obtain a plurality of independent components of each spectrum data, and a plurality of independent component sequences and ordered independent components of each spectrum data are obtained according to the wavelengths in the independent components and the absorption rates corresponding to the wavelengths.
After obtaining the spectrum data corresponding to the plurality of tetramethyl ammonium hydroxide solutions, decomposing each spectrum data by utilizing an ICA independent component analysis mode, respectively obtaining the corresponding characteristic wavelengths in each independent component in each spectrum data, and analyzing different independent components according to the DTW distances among the characteristic wavelengths to obtain an independent component sequence; the relation between the independent components and the concentrations of different substances is considered, the similarity degree between the independent components is calculated, the corresponding discrete degree of one class cluster in the independent component sequence is obtained according to the similarity degree, the distance measurement between the different class clusters is combined to obtain the splitting preference of each class cluster, and the splitting preference is brought into an ISODATA algorithm to obtain a clustering result.
Specifically, ICA independent component analysis is performed on the spectrum data to obtain a plurality of independent components of each spectrum data, specifically as follows:
for the spectrum data of any one tetramethyl ammonium hydroxide solution, ICA independent component analysis is carried out on the spectrum data to obtain a plurality of independent components of the spectrum data; in this embodiment, the number of independent components is described as 6, that is, each spectrum data has 6 independent components after ICA independent component analysis; wherein the independent components are two-dimensional data, the horizontal axes of the independent components are different wavelengths and are the same as the wavelength range of the spectrum data, the vertical axes are the absorption ratios corresponding to the different wavelengths, and each independent component can be regarded as spectrum data corresponding to one or more substances.
After the independent components corresponding to the plurality of spectrum data are obtained, because the independent components are disordered, different independent components are required to be analyzed, in order to reflect the difference of the content of substances in different tetramethyl ammonium hydroxide solutions, characteristic wavelengths in the independent components are firstly obtained to represent the substances represented by the independent components, the independent components are arranged according to the DTW distance of the represented substances, and an independent component sequence is obtained, so that the comparison of the same independent components is facilitated.
Specifically, according to the wavelength and the absorption rate corresponding to the wavelength in the independent component, a plurality of characteristic wavelengths of each independent component of each spectrum data are obtained, specifically as follows:
for any one of the individual components of any one of the spectral data, ifAnd->The corresponding independent component is in the form of +.>The wavelength is the center, the neighborhood radius is +.>The sum of the absorptances corresponding to all wavelengths in the wavelength range of (2) is greater than 1, and +.>A plurality of wavelengths as a characteristic wavelength of the independent component;
wherein the method comprises the steps ofRepresenting the +.>Absorption corresponding to the individual wavelength, < >>For a preset second value of the number,representing the independent component as +. >The wavelength is the center, the neighborhood radius is +.>In the wavelength range->Absorption corresponding to the individual wavelength, < >>。
In the case of the first embodimentThe wavelengths being on the left and right sides of the wavelength range of the independent component, no analysis of the characteristic wavelengths is performed, i.e. at +.>The wavelength is the center, the neighborhood radius is +.>The wavelength range of the individual components will be exceeded.
It should be noted that after ICA independent component analysis is performed on each spectrum data, each independent component may approximately represent spectrum data corresponding to different substances, that is, a local maximum point corresponding to each independent component has a larger probability of representing a wavelength of an absorption peak of a corresponding substance in each tetramethylammonium hydroxide solution, but when the spectrum data is decomposed, the variability of multiple independent components at the same wavelength position may occur, but when the absorption rate of the spectrum data after the actual linear combination is smaller, only selecting a local extremum point will result in partial abnormal data selection, so that further screening is needed by utilizing the absorption rate condition of the spectrum data under the corresponding wavelength.
Further, according to the characteristic wavelengths of each independent component of the spectrum data, a plurality of independent component sequences and ordered independent components of other spectrum data under each spectrum data are obtained, specifically as follows:
And recording any one spectrum data as target spectrum data, artificially numbering all independent components of the target spectrum data to obtain a plurality of independent components with an arrangement sequence, and arranging all characteristic wavelengths of a first independent component of the target spectrum data in a sequence from small to large to obtain a characteristic wavelength sequence of the first independent component of the target spectrum data.
The characteristic wavelength sequence of the first independent component of the target spectrum data is recorded as a first sequence, and the characteristic wavelength sequence with the highest similarity with the first sequence is obtained from the characteristic wavelength sequences of all independent components of any spectrum data except the target spectrum data and recorded as a similar characteristic wavelength sequence of the first sequence, wherein the specific obtaining method of the similarity is as follows: obtaining similarity between the characteristic wavelength sequence and the first sequence by using a DTW algorithm; it should be noted that, the existing method for obtaining the similarity of the two sequences by using the DTW algorithm is the DTW algorithm, and this embodiment is not described in detail.
And in the characteristic wavelength sequences of all independent components of each spectrum data except the target spectrum data, acquiring a plurality of similar characteristic wavelength sequences of the first sequence, arranging the independent components corresponding to the first sequence and the independent components corresponding to the plurality of similar characteristic wavelength sequences of the first sequence from large to small according to the similarity of the characteristic wavelength sequences to obtain a long sequence, and marking the long sequence as the independent component sequence corresponding to the first independent component of the target spectrum data, wherein the independent component corresponding to the first sequence is the first independent component of the target spectrum data at the first position of the independent component sequence, and meanwhile, arranging the independent components corresponding to the plurality of similar characteristic wavelength sequences of the first sequence into the first independent component in the belonging spectrum data, namely numbering the independent components of other spectrum data.
And marking the characteristic wavelength sequence of the second independent component of the target spectrum data as a second sequence, and acquiring the characteristic wavelength sequence with the highest similarity with the second sequence from the characteristic wavelength sequences of all independent components of any spectrum data except the target spectrum data as a similar characteristic wavelength sequence of the second sequence.
And in the characteristic wavelength sequences of all independent components of each spectrum data except the target spectrum data, acquiring a plurality of similar characteristic wavelength sequences of a second sequence, arranging the independent components corresponding to the second sequence and the independent components corresponding to the plurality of similar characteristic wavelength sequences of the second sequence from large to small according to the similarity of the characteristic wavelength sequences to obtain a long sequence, and marking the long sequence as an independent component sequence corresponding to the second independent component of the target spectrum data, wherein the independent components corresponding to the second sequence, namely the second independent component of the target spectrum data, are arranged at a first position of the independent component sequence, and meanwhile, the independent components corresponding to the plurality of similar characteristic wavelength sequences of the second sequence are arranged as second independent components in the spectrum data.
And the like, obtaining an independent component sequence corresponding to each independent component of the target spectrum data and ordered independent components of other spectrum data under the target spectrum data.
To this end, several independent component sequences and ordered independent components of each spectral data are obtained.
Step S003, when all the spectrum data are acquired to carry out iterative self-organizing clustering, a plurality of class clusters are clustered for the first time, and the matching wavelength pairs are carried out when DTW matching is carried out according to the ordered independent component of each spectrum data, the independent component of each spectrum data in each class cluster and the independent components of different spectrum data in each class cluster, so that the optimization degree that each class cluster is split into two class clusters when carrying out iterative self-organizing clustering is obtained.
It should be noted that, because the tetramethylammonium hydroxide solution may have more impurity types but less relative main components, the concentration of tetramethylammonium hydroxide solution and various impurities can be compared by using the corresponding independent components after ICA independent component analysis is performed on the spectrum data; in order to make the spectrum data of the same type gather into a cluster when clustering as much as possible, that is, some independent components of the spectrum data of the same type have a smaller degree of dispersion, it is necessary to quantify the degree of dispersion of each cluster corresponding to the sequence of independent components.
Specifically, when all spectrum data are acquired to carry out iterative self-organizing clustering, a plurality of clusters are clustered for the first time, and the distance measurement adopts the DTW distance between the spectrum data; it should be noted that, iterative self-organizing clustering is performed on all spectrum data to obtain a plurality of clusters of any one-time clustering, which is an existing method, and this embodiment is not described in detail;
Further, matching wavelength pairs when performing DTW matching according to the ordered independent components of each spectrum data, the independent components of the spectrum data in each class cluster and the independent components of different spectrum data in each class cluster, so as to obtain the similarity degree between the independent components of different spectrum data in each class cluster, wherein the similarity degree is specifically as follows:
and marking any one of the class clusters as a target class cluster.
In the method, in the process of the invention,for the p-th independent component of the i-th spectral data in the target class cluster,/for the i-th spectral data in the target class cluster>Is the p independent component of the j-th spectrum data in the target cluster, wherein +.>，/>The pearson correlation coefficient for the p-th independent component of the ith spectral data in the target class cluster and the p-th independent component of the jth spectral data in the target class cluster,/>When DTW matching is carried out on the p independent component of the ith spectrum data in the target class cluster and the p independent component of the jth spectrum data in the target class cluster, the number of the matched wavelength pairs is matched; wherein each matched wavelength pair comprises two wavelengths, one wavelength from the p-th independent component of the ith spectral data and the other wavelength from the p-th independent component of the jth spectral data;for the wavelength of the p-th independent component belonging to the i-th spectral data in the k-th matched wavelength pair, is- >For the wavelength of the p-th independent component belonging to the j-th spectral data in the k-th matched wavelength pair, is>Is the +.>First->Weight coefficient of individual components, +.>Is the +.>First->The weighting coefficients of the individual components, where +.>The sum of the individual spectral data->Weight coefficient and +.>First->The weight coefficients of the individual components can be obtained through ICA independent component analysis, and the specific obtaining method is that a mixing matrix is obtained according to ICA independent component analysis, each element in the first row in the mixing matrix is used as the weight coefficient of the corresponding independent component, which is not described in detail in this embodiment>As an exponential function based on natural constants, < +.>To take absolute value, the present embodiment uses +.>The model presents inverse proportion and normalization processing, U is the input value of the model, and an implementer can set an inverse proportion function and a normalization function according to specific implementation conditions>The similarity degree between the p independent component of the ith spectrum data in the target class cluster and the p independent component of the jth spectrum data in the target class cluster is obtained.
Further, according to the similarity between the independent components of different spectrum data in each class cluster, the discrete degree of the independent components of the spectrum data in each class cluster in the corresponding independent component sequence is obtained, which is specifically as follows:
In the method, in the process of the invention,for the number of spectral data in the target class cluster, < >>For the similarity degree of the p-th independent component of the ith spectrum data in the target class cluster and the p-th independent component of the jth spectrum data in the target class cluster, < >>To avoid superparameters with denominator 0, the present embodiment uses +.>To make a description of->The p-th independent component of the spectrum data in the target class cluster is the discrete degree of the corresponding p-th independent component sequence.
It should be noted that, the discrete degree of the target class cluster under the same dimension is obtained from the similarity degree of the independent components corresponding to each spectrum data in the target class cluster; when the similarity of the two corresponding independent components is calculated, as the corresponding independent components are obtained by decomposing the spectrum data, the absorptivity of the corresponding independent components under the corresponding characteristic wavelength is influenced by the corresponding linear combination coefficient, so that the correlation coefficient of the corresponding independent components is used as a weight, the similarity under the corresponding matched wavelength is calculated according to the matched wavelength in the DTW distance after the linear combination coefficient is used for correction, and the similarity between the two independent components can be obtained after each DTW matched wavelength pair is traversed, so that the discrete degree of the independent components of the spectrum data in the target cluster in the corresponding independent component sequence is obtained.
It should be noted that, according to the discrete degree of the single cluster under a certain independent component, the preference degree of splitting of different clusters is obtained, so that two split clusters can represent the substance concentration of a certain class, and the data discrete degree of the spectrum data corresponding to the same class should be lower, so that the distance between two clusters is increased as much as possible under the condition that the discrete degree of the two classes is smaller by using the mode of the ratio of the inter-class variance to the intra-class variance, so as to improve the accuracy of the clustering result.
Specifically, according to the discrete degree of the independent component of the spectrum data in each class cluster in the corresponding independent component sequence, the preference degree of splitting each class cluster into two class clusters when iterative self-organizing clustering is carried out is obtained, and the method specifically comprises the following steps:
when any one class cluster is subjected to iterative self-organizing clustering, the two class clusters which are split into the class cluster are respectively marked as a first sub-class cluster and a second sub-class cluster, wherein the first sub-class cluster is any one of the two class clusters, and the other class cluster is marked as the second sub-class cluster, so that no sequence requirement exists.
In the method, in the process of the invention,for the degree of discretization of the p-th independent component of the spectral data in the first sub-cluster in the sequence of the corresponding p-th independent component, >For the degree of discretization of the p-th independent component of the spectral data in the second sub-cluster in the sequence of the corresponding p-th independent component,>is the distance between the centers of the first sub-cluster and the second sub-cluster, wherein +.>The method can be obtained by an iterative self-organizing clustering algorithm, and the specific obtaining method is the existing method of the iterative self-organizing clustering algorithm, which is not described in detail in this embodiment, and the method is not described in detail>For the number of independent components per spectral data, +.>Preference for splitting the class cluster into a first sub-cluster and a second sub-cluster, +.>The normalized object is the distance between the centers of the two corresponding class clusters after each class cluster is split, which is a linear normalization function.
It should be noted that, the ratio of the inter-class variability of the two classes after splitting to the discrete degree corresponding to the two sample subsets is used as the evaluation index of the sequential splitting, when the inter-class variability is larger, i.e. the distance between the centers of the first sub-class cluster and the second sub-class cluster is larger, and the corresponding inter-class discrete degree is smaller, i.e.The cluster splitting has a better effect, so that the ISODATA algorithm tends to be stable faster, and the misjudgment risk is reduced.
So far, the preference degree of splitting each class cluster into two class clusters when iterative self-organizing clustering is carried out is obtained.
Step S004, performing iterative self-organizing clustering on all spectrum data according to the optimization degree and preset parameters of each class cluster which are split into two class clusters when performing iterative self-organizing clustering, obtaining a plurality of final class clusters, and detecting impurities of a part of the tetramethylammonium hydroxide solution corresponding to the final class clusters to obtain main impurities of the tetramethylammonium hydroxide solution corresponding to each final class cluster.
It should be noted that, by setting parameters of the iterative self-organizing clustering algorithm and splitting into two class clusters according to the preference of each class cluster when performing iterative self-organizing clustering, iterative self-organizing clustering on all spectrum data is completed, and a plurality of final class clusters are obtained.
Specifically, according to the preference degree and preset parameters of each class cluster divided into two class clusters when iterative self-organizing clustering is performed, iterative self-organizing clustering is performed on all spectrum data to obtain a plurality of final class clusters, and the specific steps are as follows:
(1) When iterative self-organizing clustering is preset to be carried out on all the spectrum data, the number of expected clustering centers is 5.
(2) When iterative self-organizing clustering is preset to be carried out on all the spectrum data, the number of the spectrum data in the minimum class cluster is 5.
(3) When iterative self-organizing clustering is preset to be carried out on all the spectrum data, the discrete degree of spectrum data distribution in the cluster is 10.
(4) When iterative self-organizing clustering is performed on all spectrum data, a splitting threshold value for splitting the class clusters is preset to be 0.9, if the preference degree of splitting the class clusters into two class clusters during iterative self-organizing clustering is larger than the splitting threshold value, splitting is performed, and otherwise, splitting is not performed.
(5) When iterative self-organizing clustering is preset to be carried out on all the spectrum data, the merging threshold value for merging different types of clusters is 8, if the distance between the centers of the different types of clusters is smaller than the merging threshold value, merging is carried out, otherwise, merging is not carried out.
(6) When iterative self-organizing clustering is preset to be carried out on all spectrum data, the maximum logarithm of class clusters which can be combined in one iteration operation is 3, namely the combining times are 3.
(7) When iterative self-organizing clustering is preset to be carried out on all the spectrum data, the maximum iterative times is 30.
And (5) performing iterative self-organizing clustering on all the spectrum data to obtain a plurality of final class clusters. It should be noted that, the specific iterative self-organizing clustering process is an existing method, and this embodiment is not described in detail.
It should be noted that different types of impurities may show different characteristics in the spectrum data, and by detecting impurities in the tetramethylammonium hydroxide solution corresponding to part of the spectrum data of the final cluster, main components of impurities in the same cluster can be obtained, so that potential problems can be found and solved in time, and the process for producing and filtering tetramethylammonium hydroxide is improved.
Specifically, the main impurities of the tetramethyl ammonium hydroxide solution corresponding to each final cluster are obtained by detecting the impurities of the part of the tetramethyl ammonium hydroxide solution corresponding to the final cluster, and the main impurities are specifically as follows:
randomly sampling and selecting N1 tetramethyl ammonium hydroxide solutions corresponding to the spectrum data in any one final cluster, respectively detecting impurities in the selected N1 tetramethyl ammonium hydroxide solutions, taking the impurity with the highest impurity content as the main impurity of the tetramethyl ammonium hydroxide solution corresponding to the final cluster, wherein N1 is a preset third value, the embodiment is described by N1=3, and an implementer can adjust according to specific implementation conditions; the main impurities of the tetramethyl ammonium hydroxide solution corresponding to the final cluster are obtained through the impurity detection of the tetramethyl ammonium hydroxide solution corresponding to the final cluster, and the purity of the tetramethyl ammonium hydroxide solution is checked and maintained regularly, so that the existence of the impurities is avoided, and good developing effect and photoetching quality can be ensured; because the final cluster contains a plurality of spectrum data, if the tetramethylammonium hydroxide solution corresponding to each spectrum data is subjected to impurity detection, the detection cost is too high, and therefore, N1 spectrum data are randomly sampled and selected in the spectrum data corresponding to any final cluster, and the N1 spectrum data are subjected to impurity detection, and because the N1 spectrum data are in the same final cluster, the main components of impurities are the same, and a small amount of tetramethylammonium hydroxide solution corresponding to the spectrum data is randomly sampled and selected to perform impurity detection, so that the tetramethylammonium hydroxide solution is used as the main impurities corresponding to the final cluster.
Through the steps, the method for detecting the filtered impurities in the production of the tetramethylammonium hydroxide is completed.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. The method for detecting the filtered impurities in the production of tetramethylammonium hydroxide is characterized by comprising the following steps of:
collecting spectrum data of a plurality of tetramethyl ammonium hydroxide solutions;
acquiring a plurality of independent components of each spectrum data, acquiring a plurality of characteristic wavelengths of each independent component of each spectrum data according to the wavelength in the independent components and the absorption rate corresponding to the wavelength, and acquiring a plurality of independent component sequences and ordered independent components of other spectrum data under each spectrum data according to a plurality of characteristic wavelengths of each independent component of the spectrum data;
when all spectrum data are obtained for iterative self-organizing clustering, a plurality of clusters are clustered for the first time, wavelength pairs are matched according to ordered independent components of each spectrum data, independent components of spectrum data in each cluster and independent components of different spectrum data in each cluster when DTW matching is carried out, so that the similarity degree between the independent components of different spectrum data in each cluster is obtained, the discrete degree of the independent components of the spectrum data in each cluster in the corresponding independent component sequence is obtained according to the similarity degree of the independent components of the spectrum data in each cluster in the corresponding independent component sequence, and the preference degree of splitting each cluster into two clusters when iterative self-organizing clustering is carried out is obtained;
According to the optimization degree and preset parameters of each class cluster divided into two class clusters when iterative self-organizing clustering is carried out, iterative self-organizing clustering is carried out on all spectrum data to obtain a plurality of final class clusters, and impurity detection is carried out on part of the tetramethylammonium hydroxide solution corresponding to the final class clusters to obtain main impurities of the tetramethylammonium hydroxide solution corresponding to each final class cluster.
2. The method for detecting the filtered impurities in the production of the tetramethylammonium hydroxide according to claim 1, wherein the obtaining the characteristic wavelengths of each independent component of each spectrum data according to the wavelength and the absorption rate corresponding to the wavelength in the independent component comprises the following specific steps:
for any one of the individual components of any one of the spectral data, ifAnd->The corresponding independent component is in the form of +.>The wavelength is the center, the neighborhood radius is +.>The sum of the absorptances corresponding to all wavelengths in the wavelength range of (2) is greater than 1, and +.>A plurality of wavelengths as a characteristic wavelength of the independent component;
wherein the method comprises the steps ofRepresenting the +.>Absorption corresponding to the individual wavelength, < >>Is a preset second value, < + >>Representing the independent component as +.>The wavelength is the center, the neighborhood radius is +. >In the wavelength range->The absorption rate corresponding to the respective wavelength is,。
3. the method for detecting the filtered impurities in the production of the tetramethylammonium hydroxide according to claim 1, wherein the steps of obtaining a plurality of independent component sequences and ordered independent components of other spectrum data in each spectrum data according to a plurality of characteristic wavelengths of each independent component of the spectrum data comprise the following specific steps:
recording any one spectrum data as target spectrum data, artificially numbering all independent components of the target spectrum data to obtain a plurality of independent components with an arrangement sequence, and arranging all characteristic wavelengths of a first independent component of the target spectrum data in a sequence from small to large to obtain a characteristic wavelength sequence of the first independent component of the target spectrum data;
the characteristic wavelength sequence of the first independent component of the target spectrum data is recorded as a first sequence, and the characteristic wavelength sequence with the highest similarity with the first sequence is obtained from the characteristic wavelength sequences of all independent components of any spectrum data except the target spectrum data and recorded as a similar characteristic wavelength sequence of the first sequence;
in the characteristic wavelength sequences of all independent components of each spectrum data except the target spectrum data, acquiring a plurality of similar characteristic wavelength sequences of a first sequence, arranging the independent components corresponding to the first sequence and the independent components corresponding to the plurality of similar characteristic wavelength sequences of the first sequence from large to small according to the similarity of the characteristic wavelength sequences to obtain a long sequence, and marking the long sequence as an independent component sequence corresponding to the first independent component of the target spectrum data, wherein the independent component corresponding to the first sequence is arranged at the first position of the independent component sequence, and meanwhile, the independent components corresponding to the plurality of similar characteristic wavelength sequences of the first sequence are arranged as the first independent component in the spectrum data to which the independent components corresponding to the first sequence belong;
The characteristic wavelength sequence of the second independent component of the target spectrum data is recorded as a second sequence, and the characteristic wavelength sequence with the highest similarity with the second sequence is obtained from the characteristic wavelength sequences of all independent components of any spectrum data except the target spectrum data and recorded as a similar characteristic wavelength sequence of the second sequence;
in the characteristic wavelength sequences of all independent components of each spectrum data except the target spectrum data, acquiring a plurality of similar characteristic wavelength sequences of a second sequence, arranging the independent components corresponding to the second sequence and the independent components corresponding to the similar characteristic wavelength sequences of the second sequence from large to small according to the similarity of the characteristic wavelength sequences to obtain a long sequence, and marking the long sequence as an independent component sequence corresponding to the second independent component of the target spectrum data, wherein the independent component corresponding to the second sequence is arranged at a first position of the independent component sequence, and meanwhile, the independent components corresponding to the similar characteristic wavelength sequences of the second sequence are arranged as a second independent component in the spectrum data;
and the like, obtaining an independent component sequence corresponding to each independent component of the target spectrum data and ordered independent components of other spectrum data under the target spectrum data.
4. The method for detecting the filtered impurities in the production of tetramethylammonium hydroxide according to claim 3, wherein the specific method for obtaining the similarity is as follows:
and obtaining the similarity between the characteristic wavelength sequence and the first sequence by using a DTW algorithm.
5. The method for detecting the filtered impurities in the production of the tetramethylammonium hydroxide according to claim 1, wherein the matching wavelength pairs in the DTW matching process according to the ordered independent component of each spectrum data, the independent component of each spectrum data in each class cluster and the independent component of different spectrum data in each class cluster, to obtain the similarity degree between the independent components of different spectrum data in each class cluster, comprises the following specific steps:
any one of the class clusters is marked as a target class cluster;
in the method, in the process of the invention,for the p-th independent component of the i-th spectral data in the target class cluster,/for the i-th spectral data in the target class cluster>For the p-th independent component of the j-th spectral data in the target class cluster,/for>The pearson correlation coefficient for the p-th independent component of the ith spectral data in the target class cluster and the p-th independent component of the jth spectral data in the target class cluster,/>When DTW matching is carried out on the p independent component of the ith spectrum data in the target class cluster and the p independent component of the jth spectrum data in the target class cluster, the number of the matched wavelength pairs is matched; / >For the wavelength of the p-th independent component belonging to the i-th spectral data in the k-th matched wavelength pair,for the wavelength of the p-th independent component belonging to the j-th spectral data in the k-th matched wavelength pair, is>Is the +.>First->Weight coefficient of individual components, +.>Is the +.>First->The weighting coefficients of the individual components, where +.>The sum of the individual spectral data->Weight coefficient and +.>First->The weight coefficients of the individual components can be obtained by ICA individual component analysis, < >>As an exponential function based on natural constants, < +.>To take absolute value, +.>The similarity degree between the p independent component of the ith spectrum data in the target class cluster and the p independent component of the jth spectrum data in the target class cluster is obtained.
6. The method for detecting the filtered impurities in the production of the tetramethylammonium hydroxide according to claim 5, wherein the step of obtaining the discrete degree of the independent component of the spectrum data in each cluster in the corresponding independent component sequence according to the similarity degree between the independent components of the different spectrum data in each cluster comprises the following specific steps:
in the method, in the process of the invention, For the number of spectral data in the target class cluster, < >>For the similarity degree of the p-th independent component of the ith spectrum data in the target class cluster and the p-th independent component of the jth spectrum data in the target class cluster, < >>To avoid hyper-parameters with denominator 0, < ->Discrete distance of the p-th independent component of spectrum data in target class cluster in corresponding p-th independent component sequenceDegree.
7. The method for detecting the filtered impurities in the production of the tetramethylammonium hydroxide according to claim 1, wherein the obtaining the preference of each cluster to be split into two clusters when the iterative self-organizing clustering is performed according to the discrete degree of the independent component of the spectrum data in each cluster in the corresponding independent component sequence comprises the following specific steps:
when any one class cluster is subjected to iterative self-organizing clustering, the two class clusters divided into the class cluster are respectively marked as a first sub-class cluster and a second sub-class cluster;
in the method, in the process of the invention,for the degree of discretization of the p-th independent component of the spectral data in the first sub-cluster in the sequence of the corresponding p-th independent component,>for the degree of discretization of the p-th independent component of the spectral data in the second sub-cluster in the sequence of the corresponding p-th independent component,>for the distance between the centers of the first sub-cluster and the second sub-cluster, < > >For the number of independent components per spectral data, +.>Preference for splitting the class cluster into a first sub-cluster and a second sub-cluster, +.>Is a linear normalization function.
8. The method for detecting the filtered impurities in the production of the tetramethylammonium hydroxide according to claim 1, wherein the step of collecting the spectrum data of a plurality of tetramethylammonium hydroxide solutions comprises the following specific steps:
a first numerical value is preset and is recorded as N, and a multispectral instrument is utilized to respectively obtain the spectrum data of N tetramethyl ammonium hydroxide solutions.
9. The method for detecting the filtered impurities in the production of tetramethyl ammonium hydroxide according to claim 1, wherein the specific method for obtaining the independent components of each spectrum data is as follows:
and (3) performing ICA independent component analysis on each spectrum data to obtain a plurality of independent components of each spectrum data.
10. The method for detecting the impurities in the tetramethylammonium hydroxide production according to claim 1, wherein the step of detecting the impurities in the tetramethylammonium hydroxide solution corresponding to the final cluster to obtain the main impurities in the tetramethylammonium hydroxide solution corresponding to each final cluster comprises the following steps:
Randomly sampling and selecting N1 tetramethyl ammonium hydroxide solutions corresponding to the spectrum data in any one final cluster, respectively detecting impurities of the selected N1 tetramethyl ammonium hydroxide solutions, taking the impurity with the highest impurity content as the main impurity of the tetramethyl ammonium hydroxide solution corresponding to the final cluster, wherein N1 is a preset third value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410009120.1A CN117556245B (en) | 2024-01-04 | 2024-01-04 | Method for detecting filtered impurities in tetramethylammonium hydroxide production |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410009120.1A CN117556245B (en) | 2024-01-04 | 2024-01-04 | Method for detecting filtered impurities in tetramethylammonium hydroxide production |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117556245A CN117556245A (en) | 2024-02-13 |
| CN117556245B true CN117556245B (en) | 2024-03-22 |
Family
ID=89818766
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410009120.1A Active CN117556245B (en) | 2024-01-04 | 2024-01-04 | Method for detecting filtered impurities in tetramethylammonium hydroxide production |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117556245B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130045543A (en) * | 2011-10-26 | 2013-05-06 | 한국과학기술원 | Nanocomposite having ionic clathrate hydrates and clay particles and preparation method thereof |
| WO2015168337A1 (en) * | 2014-04-29 | 2015-11-05 | Colorado School Of Mines | Stable metal oxide catalysts for metal oxide laser ionization-mass spectrometry |
| CN106082351A (en) * | 2016-06-01 | 2016-11-09 | 浙江大学 | The preparation method of a kind of FeOOH nanometer sheet and product thereof |
| CN109470686A (en) * | 2018-11-19 | 2019-03-15 | 北京理工大学 | A kind of laser induced breakdown spectroscopy classifying identification method based on semi-supervised learning |
| CN113227754A (en) * | 2018-08-16 | 2021-08-06 | Essenlix 公司 | Image-based assays using smart monitoring structures |
| CN117195018A (en) * | 2023-11-03 | 2023-12-08 | 南通银河水泵有限公司 | Mining water pump intelligent monitoring system based on multiple sensors |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11681839B2 (en) * | 2019-09-30 | 2023-06-20 | The Board Of Trustees Of The University Of Alabama | Systems and methods for modeling water quality |
-
2024
- 2024-01-04 CN CN202410009120.1A patent/CN117556245B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130045543A (en) * | 2011-10-26 | 2013-05-06 | 한국과학기술원 | Nanocomposite having ionic clathrate hydrates and clay particles and preparation method thereof |
| WO2015168337A1 (en) * | 2014-04-29 | 2015-11-05 | Colorado School Of Mines | Stable metal oxide catalysts for metal oxide laser ionization-mass spectrometry |
| CN106082351A (en) * | 2016-06-01 | 2016-11-09 | 浙江大学 | The preparation method of a kind of FeOOH nanometer sheet and product thereof |
| CN113227754A (en) * | 2018-08-16 | 2021-08-06 | Essenlix 公司 | Image-based assays using smart monitoring structures |
| CN109470686A (en) * | 2018-11-19 | 2019-03-15 | 北京理工大学 | A kind of laser induced breakdown spectroscopy classifying identification method based on semi-supervised learning |
| CN117195018A (en) * | 2023-11-03 | 2023-12-08 | 南通银河水泵有限公司 | Mining water pump intelligent monitoring system based on multiple sensors |
Non-Patent Citations (1)
| Title |
|---|
| "基于图像的瓦斯灰目标提取";唐艳艳;《工程科技Ⅰ辑》;20180215(第2期);第1-6章 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117556245A (en) | 2024-02-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7899625B2 (en) | Method and system for robust classification strategy for cancer detection from mass spectrometry data | |
| CN109187392B (en) | Zinc liquid trace metal ion concentration prediction method based on partition modeling | |
| CN112712108B (en) | Raman spectrum multivariate data analysis method | |
| CN111488926B (en) | Soil organic matter determination method based on optimization model | |
| US20230243744A1 (en) | Method and system for automatically detecting and reconstructing spectrum peaks in near infrared spectrum analysis of tea | |
| CN103278467A (en) | Rapid nondestructive high-accuracy method with for identifying abundance degree of nitrogen element in plant leaf | |
| CN115221927A (en) | Ultraviolet-visible spectrum dissolved organic carbon detection method | |
| CN111582387A (en) | Rock spectral feature fusion classification method and system | |
| CN115420707A (en) | Sewage near infrared spectrum chemical oxygen demand assessment method and system | |
| CN116204831A (en) | Road-to-ground analysis method based on neural network | |
| CN117309838A (en) | Industrial park water pollution tracing method based on three-dimensional fluorescence characteristic data | |
| CN114611582A (en) | Method and system for analyzing substance concentration based on near infrared spectrum technology | |
| CN108827909B (en) | Rapid soil classification method based on visible near infrared spectrum and multi-target fusion | |
| CN117556245B (en) | Method for detecting filtered impurities in tetramethylammonium hydroxide production | |
| CN112750507A (en) | Method for simultaneously detecting content of nitrate and nitrite in water based on hybrid machine learning model | |
| Wang et al. | SVM classification method of waxy corn seeds with different vitality levels based on hyperspectral imaging | |
| CN114863291B (en) | Hyperspectral image band selection method based on MCL and spectrum difference measurement | |
| CN110887798A (en) | Nonlinear full-spectrum water turbidity quantitative analysis method based on extreme random tree | |
| CN114062306B (en) | Near infrared spectrum data segmentation preprocessing method | |
| CN114184599B (en) | Single-cell Raman spectrum acquisition number estimation method, data processing method and device | |
| CN115236044A (en) | Method and device for calculating concentration of soluble organic carbon in water environment by fluorescence spectrometry | |
| CN116399836A (en) | Cross-talk fluorescence spectrum decomposition method based on alternating gradient descent algorithm | |
| CN101667253A (en) | Supervised classification method of multi-class hyperspectrum remotely sensed data | |
| CN112801172A (en) | Chinese cabbage pesticide residue qualitative analysis method based on fuzzy pattern recognition | |
| CN112326574A (en) | Spectrum wavelength selection method based on Bayesian classification |
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 |