CN115993344A - Quality monitoring and analyzing system and method for near infrared spectrum analyzer - Google Patents
Quality monitoring and analyzing system and method for near infrared spectrum analyzer Download PDFInfo
- Publication number
- CN115993344A CN115993344A CN202310287922.4A CN202310287922A CN115993344A CN 115993344 A CN115993344 A CN 115993344A CN 202310287922 A CN202310287922 A CN 202310287922A CN 115993344 A CN115993344 A CN 115993344A
- Authority
- CN
- China
- Prior art keywords
- detection
- repeatability
- analysis
- results
- content information
- 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.)
- Pending
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 59
- 238000002329 infrared spectrum Methods 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000001514 detection method Methods 0.000 claims abstract description 251
- 238000004458 analytical method Methods 0.000 claims abstract description 204
- 239000000126 substance Substances 0.000 claims abstract description 74
- 238000004364 calculation method Methods 0.000 claims abstract description 21
- 238000001228 spectrum Methods 0.000 claims description 57
- 238000010276 construction Methods 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 16
- 238000012795 verification Methods 0.000 claims description 14
- 238000012360 testing method Methods 0.000 claims description 13
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000013215 result calculation Methods 0.000 claims description 4
- 238000010183 spectrum analysis Methods 0.000 claims description 2
- 230000003595 spectral effect Effects 0.000 claims 1
- 230000000694 effects Effects 0.000 abstract description 7
- 238000012545 processing Methods 0.000 abstract description 3
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000002835 absorbance Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000004497 NIR spectroscopy Methods 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000031700 light absorption Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a quality monitoring analysis system and method of a near infrared spectrum analyzer, which relate to the technical field of data processing, and the method comprises the following steps: obtaining detection substances for quality detection, and performing near infrared spectrum scanning detection on the detection substances for M times by using a target analyzer to obtain M detection results; calculating errors between the detected content and the actual content to obtain M error parameter sets; obtaining M total error parameters through weighted calculation and obtaining the repeatability error standard deviation of the M total error parameters through calculation; and respectively carrying out accuracy analysis and repeatability analysis by using a detection accuracy analysis unit and a atlas repeatability analysis unit in the quality monitoring model, and taking the atlas repeatability analysis result and the comprehensive detection accuracy analysis result as quality detection results of the target analyzer. The invention solves the technical problems of low intelligent degree and low analysis accuracy of quality monitoring analysis of the analyzer in the prior art, and achieves the technical effect of improving the quality monitoring accuracy.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a quality monitoring and analyzing system and method of a near infrared spectrum analyzer.
Background
Near infrared spectroscopy is a three-in-one rapid analysis by comprehensively utilizing the techniques of spectroscopy, computer and chemometrics. At present, manufacturers of near infrared spectrum analyzers are numerous, the quality is uneven, the quality of the near infrared spectrum analyzers is researched and analyzed, and the near infrared spectrum analyzers have very important significance for improving the application value of the analyzers.
At present, in the face of a plurality of different near-external red spectrum analyzers, the analyzers are mainly evaluated in a mode of establishing analysis indexes such as wavelength range, resolution, light absorption accuracy and the like, so that quality analysis results are obtained. However, the obtained index values are different based on different production standards of different manufacturers only by an index judgment mode, and the analysis result cannot accurately reflect the actual quality of the analyzer. In addition, the index analysis process often uses a manual recording mode, so that the analysis period is long and the analysis efficiency is low. In the prior art, the technical problems of low intelligent degree of quality monitoring analysis and low analysis accuracy of an analyzer exist.
Disclosure of Invention
The application provides a near infrared spectrum analyzer quality monitoring analysis system and method, which are used for solving the technical problems of low intelligent degree and low analysis accuracy of analyzer quality monitoring analysis in the prior art.
In view of the above problems, the present application provides a quality monitoring analysis system and method for a near infrared spectrum analyzer.
In a first aspect of the present application, there is provided a quality monitoring analysis method of a near infrared spectrum analyzer, the method comprising:
obtaining a detection substance for detecting the quality of a target analyzer, wherein the target analyzer is a near infrared spectrum analyzer, the detection substance consists of P components with P pieces of actual content information, and P is an integer greater than 1;
performing near infrared spectrum scanning detection on the detection substance for M times by adopting the target analyzer to obtain M scanning spectrum information, wherein M is a certificate larger than 1;
inputting the M scanning spectrum information into an analysis model of the target analyzer, and obtaining the content of P components in the detection substance to obtain M detection results, wherein each detection result comprises P detection content information of the P components;
respectively calculating error parameters of P pieces of detection content information and P pieces of actual content information in the M detection results to obtain M error parameter sets;
respectively carrying out weighted calculation on error parameters in the M error parameter sets according to the size of the P actual content information to obtain M total error parameters, and calculating to obtain the repeatability error standard deviation of the M total error parameters;
Inputting the M total error parameters into a detection accuracy analysis unit in a quality monitoring model to obtain M detection accuracy analysis results, and calculating to obtain a comprehensive detection accuracy analysis result;
inputting the repeatability error standard deviation into a graph repeatability analysis unit in the quality monitoring model to obtain a graph repeatability analysis result, and combining the comprehensive detection accuracy analysis result to serve as a quality monitoring result of the target analyzer.
In a second aspect of the present application, there is provided a near infrared spectrum analyzer quality monitoring analysis system, the system comprising:
the detection substance obtaining module is used for obtaining detection substances for quality detection of a target analyzer, wherein the target analyzer is a near infrared spectrum analyzer, the detection substances consist of P components with P pieces of actual content information, and P is an integer larger than 1;
the spectrum information acquisition module is used for carrying out near infrared spectrum scanning detection on the detection substance for M times by adopting the target analyzer to acquire M scanning spectrum information, wherein M is a certificate larger than 1;
the detection result obtaining module is used for inputting the M scanning spectrum information into an analysis model of the target analyzer, obtaining the content of P components in the detection substance, and obtaining M detection results, wherein each detection result comprises P detection content information of the P components;
The error parameter acquisition module is used for respectively calculating error parameters of the P pieces of detection content information and the P pieces of actual content information in the M detection results to obtain M error parameter sets;
the standard deviation calculation module is used for respectively carrying out weighted calculation on error parameters in the M error parameter sets according to the P actual content information to obtain M total error parameters and calculating to obtain the repeatability error standard deviation of the M total error parameters;
the analysis result calculation module is used for inputting the M total error parameters into a detection accuracy analysis unit in the quality monitoring model to obtain M detection accuracy analysis results, and calculating to obtain a comprehensive detection accuracy analysis result;
the monitoring result setting module is used for inputting the repeatability error standard deviation into a graph repeatability analysis unit in the quality monitoring model to obtain a graph repeatability analysis result, and combining the comprehensive detection accuracy analysis result to serve as a quality monitoring result of the target analyzer.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
according to the method, a detection substance for quality detection of a target analyzer is obtained, the target analyzer is a near infrared spectrum analyzer, the detection substance consists of P components with P pieces of actual content information, P is an integer larger than 1, then the target analyzer is adopted to carry out M times of near infrared spectrum scanning detection on the detection substance, M pieces of scanning spectrum information are obtained, M is a certificate larger than 1, then the M pieces of scanning spectrum information are input into an analysis model of the target analyzer, the content of P components in the detection substance is obtained, M detection results are obtained, P pieces of detection content information of the P pieces of components are included in each detection result, then error parameters of the P pieces of detection content information and the P pieces of actual content information in the detection results are respectively calculated, M error parameter sets are obtained, then according to the size of the P pieces of actual content information, M total error parameters are respectively calculated in a weighting mode, M total error parameters are obtained, the M total error parameters are calculated, the M total error parameters are obtained, the M total error parameters are calculated and are input into an analysis model, the quality error of the quality detection unit is obtained, the quality detection unit is repeatedly detected, the quality error of the quality detection unit is analyzed, and the quality detection unit is obtained, and the quality error of the quality analysis is analyzed by combining the quality analysis result, and the quality analysis result is obtained. The quality of the target analyzer is comprehensively monitored, the accuracy of analysis results is guaranteed, and the technical effect of intelligent degree of analysis is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a quality monitoring and analyzing method of a near infrared spectrum analyzer according to an embodiment of the present application;
fig. 2 is a schematic flow chart of obtaining M detection results in a quality monitoring and analyzing method of a near infrared spectrum analyzer according to an embodiment of the present application;
fig. 3 is a schematic flow chart of obtaining M detection accuracy analysis results in a quality monitoring analysis method of a near infrared spectrum analyzer according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of a quality monitoring and analyzing system of a near infrared spectrum analyzer according to an embodiment of the present application.
Reference numerals illustrate: the device comprises a detection substance obtaining module 11, a spectrum information obtaining module 12, a detection result obtaining module 13, an error parameter obtaining module 14, a standard deviation calculating module 15, an analysis result calculating module 16 and a monitoring result setting module 17.
Detailed Description
The application provides a near infrared spectrum analyzer quality monitoring analysis system and a near infrared spectrum analyzer quality monitoring analysis method, which are used for solving the technical problems of low intelligent degree and low analysis accuracy of analyzer quality monitoring analysis in the prior art.
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present application based on the embodiments herein.
It should be noted that the terms "comprises" and "comprising," along with any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or server that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus, but may include other steps or modules not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
As shown in fig. 1, the present application provides a quality monitoring analysis method of a near infrared spectrum analyzer, the method comprising:
step S100: obtaining a detection substance for detecting the quality of a target analyzer, wherein the target analyzer is a near infrared spectrum analyzer, the detection substance consists of P components with P pieces of actual content information, and P is an integer greater than 1;
specifically, the present application analyzes the monitoring quality of a target analyzer by acquiring a detection substance having a quantitative analysis result, that is, determining the components of the detection substance and the actual contents corresponding to each component, and by using the detection substance as a detection target. Therefore, the quality of the near infrared spectrum analyzer can be monitored and analyzed from the dimension of practical application. The target analyzer is a near infrared spectrum analyzer, and irradiates a sample by using a near infrared light source, and then analyzes effective information carried by sample substances according to light transmitted or reflected by the sample, so that components and corresponding contents in the sample substances are rapidly and accurately detected. The detection substance consists of P components with P pieces of actual content information, namely the components in the detection substance and the corresponding contents are determined, and the components and the contents detected by the target analyzer can be compared with the reference to determine the monitoring quality of the analyzer.
Step S200: performing near infrared spectrum scanning detection on the detection substance for M times by adopting the target analyzer to obtain M scanning spectrum information, wherein M is a certificate larger than 1;
specifically, the target analyzer is used for carrying out M times of near infrared spectrum scanning detection on the detection substance under the same experimental environment, and multiple sets of analysis data are provided for subsequent analysis by carrying out repeated scanning for multiple times. The M scanning spectrum information is spectrum information obtained after the detection substance absorbs or reflects the received near infrared spectrum, and comprises spectrum characteristic frequency, wavelength information, wave number information, absorbance information and the like. The M scanning spectrum information is acquired to provide basis for the follow-up determination of the detection result of the target analyzer on the detection substance.
Step S300: inputting the M scanning spectrum information into an analysis model of the target analyzer, and obtaining the content of P components in the detection substance to obtain M detection results, wherein each detection result comprises P detection content information of the P components;
further, as shown in fig. 2, the M scan spectrum information is input into an analysis model of the target analyzer, so as to obtain the contents of multiple components in the detection substance, and obtain M detection results, where step S300 further includes:
Step S310: according to the detection substances, a plurality of similar detection substances with P components are obtained, and a plurality of content information sets are obtained;
step S320: adopting the target analyzer to perform near infrared spectrum scanning detection on the plurality of similar detection substances to obtain a plurality of scanning spectrum information;
step S330: adopting the plurality of scanning spectrum information and the plurality of content information sets as construction data, and constructing the analysis model based on supervised learning;
step S340: and inputting the M scanning spectrum information into the analysis model to obtain M detection results, wherein the M detection results comprise identification information of the M detection results.
Specifically, the detection substance contains a plurality of types and has different components, and the content of each component is also different. And extracting the detection substances with the same P components from the detection substances by taking the same components as extraction basis, setting the detection substances as the detection substances of the same type, extracting the content corresponding to the P components in each detection substance, and obtaining a plurality of content information sets. Wherein the plurality of content information sets are in one-to-one correspondence with the plurality of similar detection substances.
Specifically, the target analyzer is utilized to perform near infrared spectrum scanning detection on the plurality of similar detection substances one by one according to a certain sequence, so as to obtain the plurality of scanning spectrum information. Preferably, the scanning order is set by a worker, and is not limited herein. The plurality of scanned spectrum information reflects the scanned spectrum conditions generated by the detection substance under the detection of the target analyzer, including wavelength information, wave number information, and absorbance information, in other words, the plurality of scanned spectrum information is a side reflection of the scanning capability of the target analyzer. By constructing the analysis model by taking the plurality of scanning spectrum information and the plurality of content information sets as construction data, a model conforming to the scanning analysis capability of the target analyzer can be obtained. The analysis model takes scanning spectrum information as input data, and obtains detection results containing contents of various components in the detection substances after model analysis, and the detection results are marked on model output results.
Preferably, the analysis model is a functional model constructed by taking a convolutional neural network as a basic framework, input data is scanning spectrum information, and output data is an output result. And training the model by taking the plurality of scanning spectrum information and the plurality of content information sets as construction data until the model output result reaches convergence, so as to finish the construction of the analysis model. The M detection results are obtained by inputting M scanning spectrum information into the analysis model and performing model intelligent operation, and P detection content information corresponding to P components in each detection substance is obtained. The target of detecting the content of the components in the detection substance by the target analyzer is realized.
Step S400: respectively calculating error parameters of P pieces of detection content information and P pieces of actual content information in the M detection results to obtain M error parameter sets;
further, calculating error parameters of the P detected content information and the P actual content information in the M detected results, respectively, to obtain M error parameter sets, where step S400 in this embodiment of the present application further includes:
step S410: respectively calculating the difference values of the P pieces of detected content information and the P pieces of actual content information in the M pieces of detection results to obtain M pieces of content difference value information sets;
step S420: and respectively calculating the ratio of P content difference information to the P actual content information in the M content difference information sets to obtain the M error parameter sets.
Specifically, the difference is made between the P pieces of detected content information in the M pieces of detected results and the P pieces of actual content information, and the M pieces of content difference information sets are obtained according to difference results. The M content difference information sets reflect the deviation between the detection result and the actual content of the target analyzer, and can provide analysis data for analyzing the accuracy of the target analyzer. And comparing the content difference information with the actual content information to obtain an error parameter. The error parameter set reflects the error degree of an analysis result obtained after M times of scanning by the target analyzer. By setting the ratio as the error parameter, the larger the ratio is, the larger the error is, the smaller the ratio is, the smaller the error is, and the degree of the error is quantized and analyzed.
Step S500: respectively carrying out weighted calculation on error parameters in the M error parameter sets according to the size of the P actual content information to obtain M total error parameters, and calculating to obtain the repeatability error standard deviation of the M total error parameters;
further, according to the magnitudes of the P actual content information, weighting calculation is performed on the error parameters in the M error parameter sets, where step S500 further includes:
step S510: according to the size of the P pieces of actual content information, weight distribution is carried out, and P weight values are obtained;
step S520: and respectively carrying out weighted calculation on the P error parameters in the M error parameter sets by adopting the P weights to obtain the M total error parameters.
Specifically, the content of P components in the detection substance varies, and the accuracy of the detection result is required to be different. The components with different amounts have different degrees of influence on the quality of the detection substance, and the larger the content is, the larger the influence on the quality of the detection substance is. Therefore, by performing weight distribution according to the size of the actual content information, the influence of the error parameter corresponding to the component in P on the detection substance is calculated more accurately.
Specifically, adding and calculating the P pieces of actual content information to obtain total actual content information, dividing each piece of actual content information in the P pieces of actual content information by the total actual content information, multiplying the obtained ratio by 100, thereby obtaining a weight corresponding to each piece of actual content information, and obtaining the P weights according to a weight calculation result. And respectively carrying out weighted calculation on the error parameters in the M error parameter sets by using the P weights, multiplying the P error parameters by the corresponding weights, and adding calculation results to obtain M total error parameters. The difference between the total error parameters corresponding to the M total error parameters and the average value thereof is larger by the standard deviation of the repeatability error, which indicates that most of the total error parameters are larger than the average value, and indicates that the error generated by the repeatability experiment is larger. Therefore, the technical effect of carrying out quantization calculation on the detection error of the target analyzer is achieved.
Step S600: inputting the M total error parameters into a detection accuracy analysis unit in a quality monitoring model to obtain M detection accuracy analysis results, and calculating to obtain a comprehensive detection accuracy analysis result;
Further, as shown in fig. 3, the M total error parameters are input into a detection accuracy analysis unit in the quality monitoring model to obtain M detection accuracy analysis results, where step S600 in this embodiment of the present application further includes:
step S610: based on detection data of similar analyzers of the target analyzer, acquiring total error parameters of a plurality of samples, and performing detection accuracy grade evaluation to acquire detection accuracy analysis results of the plurality of samples;
step S620: the total error parameters of the plurality of samples and the detection accuracy analysis results of the plurality of samples are used as construction data to construct the detection accuracy analysis unit;
step S630: and respectively inputting the M total error parameters into the detection accuracy analysis unit to obtain M detection accuracy analysis results.
Further, the step S620 of the embodiment of the present application further includes:
step S621: based on the total error parameters as decision features, randomly selecting the total error parameters of the samples from the total error parameters of the samples, constructing a plurality of layers of decision nodes, and dividing and deciding the input total error parameters by each layer of decision nodes;
Step S622: obtaining a plurality of final division results of the multi-layer decision node;
step S623: and correspondingly marking the plurality of final dividing results by adopting the plurality of sample detection accuracy analysis results to obtain the detection accuracy analysis unit.
Specifically, the quality monitoring model is a functional model for intelligently analyzing the quality of the target analyzer, and comprises a detection accuracy analysis unit and a map repeatability analysis unit. The detection accuracy analysis unit takes M total error parameters as input data, takes comprehensive detection accuracy analysis results as output data, and is used for intelligently analyzing the detection accuracy of the target analyzer. The atlas repeatability analysis unit takes the repeatability error standard deviation as input data, takes the atlas repeatability analysis result as output data, and is used for carrying out intelligent analysis on the repeatability of the target analyzer.
Specifically, the manufacturer and the production model of the target analyzer are obtained, the search is based on the manufacturer and the production model, the detection data search of the similar analyzers is performed based on big data, the detection accuracy grade assessment of the target analyzer is performed according to the detection data obtained by the search, namely, the accuracy grade division is performed on the sample data according to the total error parameters of the samples and the preset detection accuracy grade, so that the detection accuracy analysis results of the samples are obtained. The detection accuracy analysis results of the samples reflect the detection accuracy grade of the target analyzer corresponding to the total error parameters of the samples.
Specifically, the total error parameter is taken as a decision feature, in other words, the total error parameter is taken as a basis for decision division. By randomly selecting a plurality of sample total error parameters from the plurality of sample total error parameters (the randomly selected number of the sample total error parameters is determined by a worker according to the precision set by the analysis unit, and is not limited herein), as a plurality of layers of decision nodes, and performing class-two division on the total error parameters input into the plurality of layers of decision nodes, namely dividing the total error parameters input into the left or the right of the decision nodes. And inputting the total error parameters of the plurality of samples into the multi-layer decision nodes, and carrying out node division to obtain a plurality of final division results. And marking each final division result in the plurality of final division results according to the plurality of sample detection accuracy analysis results. Preferably, in the marking process, each final division result is marked according to the sample detection accuracy analysis result in the division result, so as to obtain the detection accuracy analysis unit.
Specifically, the M total error parameters are input into the detection accuracy analysis unit, the M total error parameters are divided through different decision nodes, corresponding division results are obtained, and then the division results are marked and determined according to marking information of a plurality of final division results in the detection accuracy analysis unit, so that the M detection accuracy analysis results are obtained according to the marking. And carrying out average value calculation on the obtained M detection accuracy analysis results so as to obtain the comprehensive detection accuracy analysis results, synthesizing errors of multiple measurements, and carrying out reliable evaluation on the detection accuracy of the target analyzer. Therefore, the target for intelligent analysis of the detection accuracy of the target analyzer is realized.
Step S700: inputting the repeatability error standard deviation into a graph repeatability analysis unit in the quality monitoring model to obtain a graph repeatability analysis result, and combining the comprehensive detection accuracy analysis result to serve as a quality monitoring result of the target analyzer.
Further, inputting the repeatability error standard deviation into a graph repeatability analysis unit in the quality monitoring model to obtain a graph repeatability analysis result, where step S700 of the embodiment of the present application further includes:
step S710: based on detection data of similar analyzers of the target analyzer, acquiring a plurality of repeatability error standard deviations, and evaluating the repeatability grades of the patterns to acquire a plurality of pattern repeatability analysis results;
step S720: marking and dividing the data of the multiple repeatability error standard deviations and the multiple atlas repeatability analysis results to obtain a training set, a verification set and a test set;
step S730: constructing and training the atlas repeatability analysis unit with accuracy meeting preset requirements based on the BP neural network by adopting a training set, a verification set and a test set;
step S740: inputting the repeatability error standard deviation into the graph repeatability analysis unit to obtain the graph repeatability analysis result.
Specifically, a plurality of repeatability error standard deviations of the similar analyzers are obtained through calculation according to the detection data, and the spectrum repeatability of the analyzers is subjected to grade assessment according to the standard deviation, so that a plurality of spectrum repeatability analysis results are obtained, wherein the plurality of spectrum repeatability analysis results reflect the spectrum repeatability conditions of the original spectrum and the derivative spectrum of the same sample repeatedly scanned by the analyzers.
Specifically, the multiple atlas repeatability analysis results are subjected to data labeling and used as supervision data in the unit training process, and then the multiple repeatability error standard deviations and the multiple atlas repeatability analysis results are divided into a training set, a verification set and a test set according to a certain division proportion, preferably, the division proportion is set by a worker, the proportion of the training set is larger than that of the verification set test set, and the specific proportion is not limited. Optionally, the ratio of training set, validation set and test set is 4:3:3.
Specifically, training the atlas repeatability analysis unit constructed based on the BP neural network as a framework by utilizing the training set, and supervising by taking a plurality of marked atlas repeatability analysis results as supervision data until the unit is trained to be converged. And further, verifying the converged pattern repeatability analysis unit by using a verification set, inputting a plurality of repeatability error standard deviations in the verification set into the unit to obtain a plurality of verification pattern repeatability analysis results, comparing and verifying the plurality of verification pattern repeatability analysis results with the plurality of pattern repeatability analysis results in the verification set, and if the successful verification matching rate reaches more than 70%, verifying to pass, wherein the pass shows that the qualification of the pattern repeatability analysis unit trained to be converged is passed.
Specifically, the test set is input into the verified atlas repeatability analysis unit, the adaptation degree of the analysis unit to data is tested, and when the test speed and the accuracy meet the requirements, the atlas repeatability analysis unit is obtained. And inputting the repeatability error standard deviation into a graph repeatability analysis unit in the quality monitoring model to obtain an output graph repeatability analysis result, analyzing and verifying the graph repeatability of the target analyzer, and combining the comprehensive detection accuracy analysis result to obtain the quality monitoring result of the target analyzer analyzed from the graph repeatability and the detection accuracy. The technical effect of accurately and efficiently analyzing the quality of the near infrared spectrum analyzer is achieved.
In summary, the embodiments of the present application have at least the following technical effects:
according to the method, the monitoring quality of the near infrared spectrum analyzer is analyzed from the dimension of practical application, so that components and detection substances with the components accounting for the content are determined in advance, the near infrared spectrum analyzer is utilized to conduct multiple scanning detection on the detection substances, the detection results are compared with the components and the component content corresponding to the detection substances, the intelligent analysis model, namely the quality monitoring model is utilized to conduct detection accuracy and map repeatability analysis on the detection results, and the technical effects of improving the intelligent degree and analysis efficiency of quality monitoring analysis of the analyzer are achieved.
Example two
Based on the same inventive concept as the quality monitoring and analyzing method of a near infrared spectrum analyzer in the foregoing embodiments, as shown in fig. 4, the present application provides a quality monitoring and analyzing system of a near infrared spectrum analyzer, and the system and method embodiments in the embodiments of the present application are based on the same inventive concept. Wherein the system comprises:
a detection substance obtaining module 11, wherein the detection substance obtaining module 11 is used for obtaining a detection substance for quality detection of a target analyzer, the target analyzer is a near infrared spectrum analyzer, the detection substance is composed of P components with P pieces of actual content information, and P is an integer greater than 1;
a spectrum information obtaining module 12, where the spectrum information obtaining module 12 is configured to perform M times of near infrared spectrum scanning detection on the detection material by using the target analyzer, to obtain M scanning spectrum information, where M is a certificate greater than 1;
the detection result obtaining module 13 is configured to input the M pieces of scanning spectrum information into an analysis model of the target analyzer, obtain contents of P components in the detection substance, and obtain M detection results, where each detection result includes P pieces of detection content information of the P components;
The error parameter obtaining module 14, the error parameter obtaining module 14 is configured to calculate error parameters of the P pieces of detected content information and the P pieces of actual content information in the M pieces of detection results, respectively, to obtain M error parameter sets;
the standard deviation calculation module 15 is configured to perform weighted calculation on the error parameters in the M error parameter sets according to the P actual content information, obtain M total error parameters, and calculate a repetitive error standard deviation of the M total error parameters;
the analysis result calculation module 16, wherein the analysis result calculation module 16 is configured to input the M total error parameters into a detection accuracy analysis unit in a quality monitoring model, obtain M detection accuracy analysis results, and calculate to obtain a comprehensive detection accuracy analysis result;
the monitoring result setting module 17 is configured to input the repeatability error standard deviation into a spectrum repeatability analysis unit in the quality monitoring model, obtain a spectrum repeatability analysis result, and combine the comprehensive detection accuracy analysis result to serve as a quality monitoring result of the target analyzer.
Further, the system further comprises:
the content information acquisition module is used for acquiring a plurality of similar detection substances with P components according to the detection substances and acquiring a plurality of content information sets;
the scanning spectrum information acquisition module is used for carrying out near infrared spectrum scanning detection on the plurality of similar detection substances by adopting the target analyzer to obtain a plurality of scanning spectrum information;
the analysis model construction module is used for adopting the plurality of scanning spectrum information and the plurality of content information sets as construction data and constructing the analysis model based on supervised learning;
the detection result acquisition module is used for inputting the M scanning spectrum information into the analysis model to obtain M detection results, and the M detection results comprise identification information of the M detection results.
Further, the system further comprises:
the content difference information obtaining module is used for respectively calculating the difference values of the P pieces of detected content information and the P pieces of actual content information in the M detection results to obtain M content difference information sets;
The error parameter set obtaining module is used for respectively calculating the ratio of the P content difference information to the P actual content information in the M content difference information sets to obtain the M error parameter sets.
Further, the system further comprises:
the weight obtaining module is used for carrying out weight distribution according to the size of the P pieces of actual content information to obtain P weight values;
and the total error parameter obtaining module is used for respectively carrying out weighted calculation on the P error parameters in the M error parameter sets by adopting the P weights to obtain the M total error parameters.
Further, the system further comprises:
the accuracy grade assessment module is used for acquiring total error parameters of a plurality of samples based on detection data of similar analyzers of the target analyzer, and carrying out detection accuracy grade assessment to acquire detection accuracy analysis results of the plurality of samples;
the accuracy analysis unit construction module is used for constructing the detection accuracy analysis unit by adopting the total error parameters of the plurality of samples and the detection accuracy analysis results of the plurality of samples as construction data;
The accuracy analysis result obtaining module is used for respectively inputting the M total error parameters into the detection accuracy analysis unit to obtain the M detection accuracy analysis results.
Further, the system further comprises:
the multi-layer decision node construction module is used for randomly selecting sample total error parameters from the plurality of sample total error parameters based on the total error parameters as decision characteristics to construct multi-layer decision nodes, and each layer of decision nodes carries out division decision on the input total error parameters;
the division result acquisition module is used for acquiring a plurality of final division results of the multi-layer decision node;
the analysis unit obtaining module is used for adopting the plurality of sample detection accuracy analysis results and correspondingly marking the plurality of final division results to obtain the detection accuracy analysis unit.
Further, the system further comprises:
the repeatability analysis result obtaining module is used for obtaining a plurality of repeatability error standard deviations based on detection data of the similar analyzers of the target analyzer, and carrying out spectrogram repeatability grade assessment to obtain a plurality of spectrogram repeatability analysis results;
The marking and dividing module is used for marking and dividing the data of the multiple repeatability error standard deviations and the multiple atlas repeatability analysis results to obtain a training set, a verification set and a test set;
the repeatability analysis unit construction module is used for constructing and training the atlas repeatability analysis unit with accuracy meeting preset requirements based on the BP neural network by adopting a training set, a verification set and a test set;
the spectrum analysis result obtaining module is used for inputting the repeatability error standard deviation into the spectrum repeatability analysis unit to obtain the spectrum repeatability analysis result.
It should be noted that the sequence of the embodiments of the present application is merely for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
The foregoing description of the preferred embodiments of the present application is not intended to limit the invention to the particular embodiments of the present application, but to limit the scope of the invention to the particular embodiments of the present application.
The specification and drawings are merely exemplary of the application and are to be regarded as covering any and all modifications, variations, combinations, or equivalents that are within the scope of the application. It will be apparent to those skilled in the art that various modifications and variations can be made in the present application without departing from the scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the present application and the equivalents thereof, the present application is intended to cover such modifications and variations.
Claims (8)
1. A near infrared spectrum analyzer quality monitoring analysis system, the system comprising:
the detection substance obtaining module is used for obtaining detection substances for quality detection of a target analyzer, wherein the target analyzer is a near infrared spectrum analyzer, the detection substances consist of P components with P pieces of actual content information, and P is an integer larger than 1;
the spectrum information acquisition module is used for carrying out near infrared spectrum scanning detection on the detection substance for M times by adopting the target analyzer to acquire M scanning spectrum information, wherein M is a certificate larger than 1;
The detection result obtaining module is used for inputting the M scanning spectrum information into an analysis model of the target analyzer, obtaining the content of P components in the detection substance, and obtaining M detection results, wherein each detection result comprises P detection content information of the P components;
the error parameter acquisition module is used for respectively calculating error parameters of the P pieces of detection content information and the P pieces of actual content information in the M detection results to obtain M error parameter sets;
the standard deviation calculation module is used for respectively carrying out weighted calculation on error parameters in the M error parameter sets according to the P actual content information to obtain M total error parameters and calculating to obtain the repeatability error standard deviation of the M total error parameters;
the analysis result calculation module is used for inputting the M total error parameters into a detection accuracy analysis unit in the quality monitoring model to obtain M detection accuracy analysis results, and calculating to obtain a comprehensive detection accuracy analysis result;
the monitoring result setting module is used for inputting the repeatability error standard deviation into a graph repeatability analysis unit in the quality monitoring model to obtain a graph repeatability analysis result, and combining the comprehensive detection accuracy analysis result to serve as a quality monitoring result of the target analyzer.
2. The system of claim 1, wherein inputting the M scan spectral information into an analysis model of the target analyzer, obtaining the content of the plurality of components in the test substance, obtaining M test results, comprises:
the content information acquisition module is used for acquiring a plurality of similar detection substances with P components according to the detection substances and acquiring a plurality of content information sets;
the scanning spectrum information acquisition module is used for carrying out near infrared spectrum scanning detection on the plurality of similar detection substances by adopting the target analyzer to obtain a plurality of scanning spectrum information;
the analysis model construction module is used for adopting the plurality of scanning spectrum information and the plurality of content information sets as construction data and constructing the analysis model based on supervised learning;
the detection result acquisition module is used for inputting the M scanning spectrum information into the analysis model to obtain M detection results, and the M detection results comprise identification information of the M detection results.
3. The system of claim 1, wherein calculating error parameters of the P detected content information and the P actual content information in the M detected results, respectively, to obtain M error parameter sets, includes:
the content difference information obtaining module is used for respectively calculating the difference values of the P pieces of detected content information and the P pieces of actual content information in the M detection results to obtain M content difference information sets;
the error parameter set obtaining module is used for respectively calculating the ratio of the P content difference information to the P actual content information in the M content difference information sets to obtain the M error parameter sets.
4. The system of claim 1, wherein weighting the error parameters in the M error parameter sets according to the magnitudes of the P actual content information, respectively, comprises:
the weight obtaining module is used for carrying out weight distribution according to the size of the P pieces of actual content information to obtain P weight values;
and the total error parameter obtaining module is used for respectively carrying out weighted calculation on the P error parameters in the M error parameter sets by adopting the P weights to obtain the M total error parameters.
5. The system of claim 1, wherein inputting the M total error parameters into a detection accuracy analysis unit in a quality monitoring model to obtain M detection accuracy analysis results comprises:
the accuracy grade assessment module is used for acquiring total error parameters of a plurality of samples based on detection data of similar analyzers of the target analyzer, and carrying out detection accuracy grade assessment to acquire detection accuracy analysis results of the plurality of samples;
the accuracy analysis unit construction module is used for constructing the detection accuracy analysis unit by adopting the total error parameters of the plurality of samples and the detection accuracy analysis results of the plurality of samples as construction data;
the accuracy analysis result obtaining module is used for respectively inputting the M total error parameters into the detection accuracy analysis unit to obtain the M detection accuracy analysis results.
6. The system of claim 5, wherein constructing the detection accuracy analysis unit using the plurality of sample total error parameters and the plurality of sample detection accuracy analysis results as construction data comprises:
The multi-layer decision node construction module is used for randomly selecting sample total error parameters from the plurality of sample total error parameters based on the total error parameters as decision characteristics to construct multi-layer decision nodes, and each layer of decision nodes carries out division decision on the input total error parameters;
the division result acquisition module is used for acquiring a plurality of final division results of the multi-layer decision node;
the analysis unit obtaining module is used for adopting the plurality of sample detection accuracy analysis results and correspondingly marking the plurality of final division results to obtain the detection accuracy analysis unit.
7. The system of claim 1, wherein inputting the repeatability error standard deviation into a graph repeatability analysis unit within the quality monitoring model, obtaining a graph repeatability analysis result, comprises:
the repeatability analysis result obtaining module is used for obtaining a plurality of repeatability error standard deviations based on detection data of the similar analyzers of the target analyzer, and carrying out spectrogram repeatability grade assessment to obtain a plurality of spectrogram repeatability analysis results;
The marking and dividing module is used for marking and dividing the data of the multiple repeatability error standard deviations and the multiple atlas repeatability analysis results to obtain a training set, a verification set and a test set;
the repeatability analysis unit construction module is used for constructing and training the atlas repeatability analysis unit with accuracy meeting preset requirements based on the BP neural network by adopting a training set, a verification set and a test set;
the spectrum analysis result obtaining module is used for inputting the repeatability error standard deviation into the spectrum repeatability analysis unit to obtain the spectrum repeatability analysis result.
8. A near infrared spectrum analyzer quality monitoring analysis method, the method comprising:
obtaining a detection substance for detecting the quality of a target analyzer, wherein the target analyzer is a near infrared spectrum analyzer, the detection substance consists of P components with P pieces of actual content information, and P is an integer greater than 1;
performing near infrared spectrum scanning detection on the detection substance for M times by adopting the target analyzer to obtain M scanning spectrum information, wherein M is a certificate larger than 1;
Inputting the M scanning spectrum information into an analysis model of the target analyzer, and obtaining the content of P components in the detection substance to obtain M detection results, wherein each detection result comprises P detection content information of the P components;
respectively calculating error parameters of P pieces of detection content information and P pieces of actual content information in the M detection results to obtain M error parameter sets;
respectively carrying out weighted calculation on error parameters in the M error parameter sets according to the size of the P actual content information to obtain M total error parameters, and calculating to obtain the repeatability error standard deviation of the M total error parameters;
inputting the M total error parameters into a detection accuracy analysis unit in a quality monitoring model to obtain M detection accuracy analysis results, and calculating to obtain a comprehensive detection accuracy analysis result;
inputting the repeatability error standard deviation into a graph repeatability analysis unit in the quality monitoring model to obtain a graph repeatability analysis result, and combining the comprehensive detection accuracy analysis result to serve as a quality monitoring result of the target analyzer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310287922.4A CN115993344A (en) | 2023-03-23 | 2023-03-23 | Quality monitoring and analyzing system and method for near infrared spectrum analyzer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310287922.4A CN115993344A (en) | 2023-03-23 | 2023-03-23 | Quality monitoring and analyzing system and method for near infrared spectrum analyzer |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115993344A true CN115993344A (en) | 2023-04-21 |
Family
ID=85995344
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310287922.4A Pending CN115993344A (en) | 2023-03-23 | 2023-03-23 | Quality monitoring and analyzing system and method for near infrared spectrum analyzer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115993344A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116205543A (en) * | 2023-05-04 | 2023-06-02 | 张家港广大特材股份有限公司 | Method and system for detecting quality of metallurgical steel by combining feedback |
| CN116777861A (en) * | 2023-06-15 | 2023-09-19 | 昆山市金康电子有限公司 | Marking quality detection method and system for laser engraving machine |
| CN117554885A (en) * | 2023-12-13 | 2024-02-13 | 浙江万昌电力设备有限公司 | Intelligent metering box operation abnormality detection method and system |
| CN117554354A (en) * | 2024-01-12 | 2024-02-13 | 江苏诺恩作物科学股份有限公司 | Spectral detection analysis method and processing system for thiabendazole wastewater sample |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108960193A (en) * | 2018-07-24 | 2018-12-07 | 中北大学 | A kind of across component infrared spectroscopy model transplantations method based on transfer learning |
| CN110907393A (en) * | 2019-11-22 | 2020-03-24 | 黑龙江八一农垦大学 | Method and device for detecting saline-alkali stress degree of plants |
| CN114878509A (en) * | 2022-05-07 | 2022-08-09 | 浙江大学 | Standard sample-free transfer method of tobacco near-infrared quantitative analysis model |
| CN115239725A (en) * | 2022-09-22 | 2022-10-25 | 张家港大裕橡胶制品有限公司 | Visual monitoring method and system for rubber dipping coating of rubber glove |
-
2023
- 2023-03-23 CN CN202310287922.4A patent/CN115993344A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108960193A (en) * | 2018-07-24 | 2018-12-07 | 中北大学 | A kind of across component infrared spectroscopy model transplantations method based on transfer learning |
| CN110907393A (en) * | 2019-11-22 | 2020-03-24 | 黑龙江八一农垦大学 | Method and device for detecting saline-alkali stress degree of plants |
| CN114878509A (en) * | 2022-05-07 | 2022-08-09 | 浙江大学 | Standard sample-free transfer method of tobacco near-infrared quantitative analysis model |
| CN115239725A (en) * | 2022-09-22 | 2022-10-25 | 张家港大裕橡胶制品有限公司 | Visual monitoring method and system for rubber dipping coating of rubber glove |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116205543A (en) * | 2023-05-04 | 2023-06-02 | 张家港广大特材股份有限公司 | Method and system for detecting quality of metallurgical steel by combining feedback |
| CN116205543B (en) * | 2023-05-04 | 2023-10-31 | 张家港广大特材股份有限公司 | Method and system for detecting quality of metallurgical steel by combining feedback |
| CN116777861A (en) * | 2023-06-15 | 2023-09-19 | 昆山市金康电子有限公司 | Marking quality detection method and system for laser engraving machine |
| CN116777861B (en) * | 2023-06-15 | 2023-11-14 | 昆山市金康电子有限公司 | Marking quality detection method and system for laser engraving machine |
| CN117554885A (en) * | 2023-12-13 | 2024-02-13 | 浙江万昌电力设备有限公司 | Intelligent metering box operation abnormality detection method and system |
| CN117554885B (en) * | 2023-12-13 | 2024-05-07 | 浙江万昌电力设备有限公司 | Intelligent metering box operation abnormality detection method and system |
| CN117554354A (en) * | 2024-01-12 | 2024-02-13 | 江苏诺恩作物科学股份有限公司 | Spectral detection analysis method and processing system for thiabendazole wastewater sample |
| CN117554354B (en) * | 2024-01-12 | 2024-03-15 | 江苏诺恩作物科学股份有限公司 | Spectral detection analysis method and processing system for thiabendazole wastewater sample |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115993344A (en) | Quality monitoring and analyzing system and method for near infrared spectrum analyzer | |
| CN101308086B (en) | Fruit internal quality on-line checking apparatus based on near infrared spectra technology | |
| CN104062257B (en) | A kind of based on the method for general flavone content near infrared ray solution | |
| CN107703097B (en) | Method for constructing model for rapidly predicting crude oil property by using near-infrared spectrometer | |
| CN107817223A (en) | The construction method of quick nondestructive real-time estimate oil property model and its application | |
| CN102590129B (en) | Method for detecting content of amino acid in peanuts by near infrared method | |
| CN101520412A (en) | Near infrared spectrum analyzing method based on isolated component analysis and genetic neural network | |
| CN104897607A (en) | Food modeling and rapid detecting integration method and system adopting portable NIRS (near infrared spectroscopy) | |
| CN101825567A (en) | Screening method for near infrared spectrum wavelength and Raman spectrum wavelength | |
| CN109669023A (en) | A kind of soil attribute prediction technique based on Multi-sensor Fusion | |
| CN102879353A (en) | Near infrared detection method for contents of protein components in peanut | |
| CN109324013A (en) | A method of it is quickly analyzed using Gaussian process regression model building oil property near-infrared | |
| CN102252971A (en) | Rapid detection method for mango hardness | |
| EP0954744A1 (en) | Calibration system for spectrographic analyzing instruments | |
| CN106770003A (en) | Wood Identification Method and system based on near-infrared spectrum technique | |
| CN111693487A (en) | Fruit sugar degree detection method and system based on genetic algorithm and extreme learning machine | |
| CN104596979A (en) | Method for measuring cellulose of reconstituted tobacco by virtue of near infrared reflectance spectroscopy technique | |
| CN104596975A (en) | Method for measuring lignin of reconstituted tobacco by paper-making process by virtue of near infrared reflectance spectroscopy technique | |
| CN111999258B (en) | Spectral baseline correction-oriented weighting modeling local optimization method | |
| CN112179871A (en) | Method for nondestructive detection of caprolactam content in sauce food | |
| CN111879709A (en) | Method and device for detecting spectral reflectivity of lake water body | |
| CN104596976A (en) | Method for determining protein of paper-making reconstituted tobacco through ear infrared reflectance spectroscopy technique | |
| CN104316492A (en) | Method for near-infrared spectrum measurement of protein content in potato tuber | |
| CN110376154A (en) | Fruit online test method and system based on spectrum correction | |
| CN108120694B (en) | Multi-element correction method and system for chemical component analysis of sun-cured red tobacco |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20230421 |