CN115128036A - Handheld near infrared spectrum detection method and device - Google Patents
Handheld near infrared spectrum detection method and device Download PDFInfo
- Publication number
- CN115128036A CN115128036A CN202210842698.6A CN202210842698A CN115128036A CN 115128036 A CN115128036 A CN 115128036A CN 202210842698 A CN202210842698 A CN 202210842698A CN 115128036 A CN115128036 A CN 115128036A
- Authority
- CN
- China
- Prior art keywords
- sample
- infrared
- model
- spectrum
- infrared spectrum
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 61
- 238000002329 infrared spectrum Methods 0.000 title claims abstract description 59
- 238000001228 spectrum Methods 0.000 claims abstract description 103
- 238000000034 method Methods 0.000 claims abstract description 88
- 239000000463 material Substances 0.000 claims abstract description 53
- 238000007689 inspection Methods 0.000 claims abstract description 23
- 238000007781 pre-processing Methods 0.000 claims abstract description 12
- 239000000126 substance Substances 0.000 claims abstract description 8
- 239000000523 sample Substances 0.000 claims description 154
- 238000007789 sealing Methods 0.000 claims description 30
- 230000003595 spectral effect Effects 0.000 claims description 30
- 239000013307 optical fiber Substances 0.000 claims description 23
- 230000008569 process Effects 0.000 claims description 18
- 238000009499 grossing Methods 0.000 claims description 17
- 238000012937 correction Methods 0.000 claims description 14
- 238000004497 NIR spectroscopy Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000012706 support-vector machine Methods 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 6
- 238000002835 absorbance Methods 0.000 claims description 5
- 239000002245 particle Substances 0.000 claims description 5
- 239000000835 fiber Substances 0.000 claims description 4
- 229910052751 metal Inorganic materials 0.000 claims description 3
- 239000002184 metal Substances 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 239000012852 risk material Substances 0.000 claims description 3
- 230000013011 mating Effects 0.000 claims 1
- 239000003814 drug Substances 0.000 abstract description 7
- 229940079593 drug Drugs 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 16
- 238000004422 calculation algorithm Methods 0.000 description 14
- 238000003860 storage Methods 0.000 description 13
- 230000006854 communication Effects 0.000 description 12
- 238000004891 communication Methods 0.000 description 12
- 238000012795 verification Methods 0.000 description 10
- 230000005540 biological transmission Effects 0.000 description 9
- 238000005259 measurement Methods 0.000 description 9
- 238000013461 design Methods 0.000 description 8
- 230000003993 interaction Effects 0.000 description 8
- 238000012545 processing Methods 0.000 description 8
- 230000001681 protective effect Effects 0.000 description 8
- 238000005457 optimization Methods 0.000 description 7
- 238000012850 discrimination method Methods 0.000 description 6
- 239000002994 raw material Substances 0.000 description 5
- 238000011161 development Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000002349 favourable effect Effects 0.000 description 4
- WFKWXMTUELFFGS-UHFFFAOYSA-N tungsten Chemical compound [W] WFKWXMTUELFFGS-UHFFFAOYSA-N 0.000 description 4
- 229910052721 tungsten Inorganic materials 0.000 description 4
- 239000010937 tungsten Substances 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 239000010405 anode material Substances 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000007175 bidirectional communication Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000007621 cluster analysis Methods 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 229910052736 halogen Inorganic materials 0.000 description 2
- 150000002367 halogens Chemical class 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000005022 packaging material Substances 0.000 description 2
- 238000012856 packing Methods 0.000 description 2
- JBRZTFJDHDCESZ-UHFFFAOYSA-N AsGa Chemical compound [As]#[Ga] JBRZTFJDHDCESZ-UHFFFAOYSA-N 0.000 description 1
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 description 1
- HBBGRARXTFLTSG-UHFFFAOYSA-N Lithium ion Chemical compound [Li+] HBBGRARXTFLTSG-UHFFFAOYSA-N 0.000 description 1
- ATJFFYVFTNAWJD-UHFFFAOYSA-N Tin Chemical compound [Sn] ATJFFYVFTNAWJD-UHFFFAOYSA-N 0.000 description 1
- RTAQQCXQSZGOHL-UHFFFAOYSA-N Titanium Chemical compound [Ti] RTAQQCXQSZGOHL-UHFFFAOYSA-N 0.000 description 1
- 238000000862 absorption spectrum Methods 0.000 description 1
- 239000005030 aluminium foil Substances 0.000 description 1
- 230000003139 buffering effect Effects 0.000 description 1
- 239000010406 cathode material Substances 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 230000003760 hair shine Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 229910052738 indium Inorganic materials 0.000 description 1
- APFVFJFRJDLVQX-UHFFFAOYSA-N indium atom Chemical compound [In] APFVFJFRJDLVQX-UHFFFAOYSA-N 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
- 229910052744 lithium Inorganic materials 0.000 description 1
- 229910001416 lithium ion Inorganic materials 0.000 description 1
- GELKBWJHTRAYNV-UHFFFAOYSA-K lithium iron phosphate Chemical compound [Li+].[Fe+2].[O-]P([O-])([O-])=O GELKBWJHTRAYNV-UHFFFAOYSA-K 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 239000004033 plastic Substances 0.000 description 1
- 229920003023 plastic Polymers 0.000 description 1
- 229920003229 poly(methyl methacrylate) Polymers 0.000 description 1
- 229920000642 polymer Polymers 0.000 description 1
- 239000004926 polymethyl methacrylate Substances 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 238000002834 transmittance Methods 0.000 description 1
- 239000012780 transparent material Substances 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
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
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/359—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using near infrared light
-
- 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
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- 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
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/3563—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light for analysing solids; Preparation of samples therefor
Landscapes
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention belongs to the field of rapid detection and release of raw and auxiliary materials of medicines, and provides a handheld near infrared spectrum detection method and a handheld near infrared spectrum detection device, wherein the handheld near infrared spectrum detection method comprises the following steps: acquiring infrared spectrum data of a sample and preprocessing the infrared spectrum data; performing band selection based on the infrared spectrum data of the preprocessed sample to obtain the infrared spectrum data of the sample to be detected; automatically establishing a near-infrared qualitative inspection model based on infrared spectrum data of a sample to be detected; judging a near-infrared qualitative inspection model by adopting a multi-stage release strategy, and judging whether the near-infrared qualitative inspection model is correct or not; if the result is correct, checking according to the near-infrared qualitative checking model, and storing the checking result into a material database; if the error occurs, the sample is not stored in the material database, and the sample is subjected to chemical detection; the method adopts an automatic multi-level release strategy to avoid the risk of a single method release judgment error, and solves the problem that the traditional modeling method needs a large amount of manpower to collect the spectrums of various samples.
Description
Technical Field
The invention belongs to the technical field of rapid detection and release of raw and auxiliary materials of medicines, and particularly relates to a handheld near infrared spectrum detection method and device.
Background
The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.
In the pharmaceutical industry, near infrared spectroscopy is used for the identification and analysis of raw materials before the raw materials are put in and accepted, and has the advantage that the raw materials can be directly analyzed through glass or polymer packages. The traditional near infrared spectrum detector has the problems of complex equipment structure, large occupied space, inconvenient carrying and the like, and is not beneficial to application and popularization in the aspect of detection of pharmaceutical raw and auxiliary materials. Compared with a special instrument, the portable near-infrared spectrometer is smaller in size, lighter in weight and wider in application range to environmental temperature, humidity and earthquake resistance. Sensing, transmission and analysis of real-time data information can be accelerated through cloud computing. The existing portable instrument is mainly handheld, the main types of the existing portable instrument are cylindrical and pistol, and the existing type detector is mostly designed for non-contact detection, so that the problems of incomplete acquired spectral information, high requirement on operator operation and the like are caused. The invention realizes the optical fiber exposure design, can directly contact with a detection sample, and avoids the problems of sample pollution or incomplete detection information reception and the like caused by improper control of the detection distance. The anti-skid design of the detector shell can better protect the instrument, the ABS material with light density and ideal comprehensive performance is selected as the material, and the purposes of portability and accurate detection are achieved.
In order to realize nondestructive inspection, the near-infrared spectrometer can detect a sample to be detected through the outer packaging material. The packaging material of part medicine raw and auxiliary materials can directly detect for transparent material, but the extranal packing of part medicine raw and auxiliary materials is non-transparent packing such as tinfoil aluminium foil, must realize detecting through the mode of certain distance of interval or direct contact material. The existing primary and secondary material line detection instrument for pharmaceutical factories is mainly non-contact detection, the detection process ensures that a probe and a detection sample are separated by a certain distance and cannot contact the sample, the optical fiber probe is prevented from being polluted, and the operation difficulty coefficient is high. Therefore, data collection is difficult, the collection success rate is relatively low, and the optical fiber probe after being detected is difficult to clean. In order to ensure that the established model has better accuracy and minimize the interference of errors, the optical fiber probe is inserted into a sample for direct measurement when the model is established. The invention avoids the problems by configuring the disposable probe sleeve for the optical fiber probe, has low manufacturing cost and convenient use, and is very suitable for detecting mass raw and auxiliary materials in pharmaceutical factories.
Aiming at the identification of the raw materials and auxiliary materials at present, most of handheld near-infrared instruments in the market mostly adopt a correlation coefficient single method to judge the attribution of a sample to be detected, but the near-infrared spectrum has no obvious characteristic peak, the spectra of different materials of the same type are very similar, and a single release method has certain risk.
Disclosure of Invention
In order to solve the problems, the invention provides a handheld near infrared spectrum detection method and a handheld near infrared spectrum detection device, and the automatic multi-stage release strategy avoids the risk of single-method release judgment errors, and is favorable for application and popularization in the aspect of detection of raw and auxiliary materials of medicines.
According to some embodiments, a first aspect of the present invention provides a handheld near infrared spectrum detection method, which adopts the following technical solutions:
a handheld near infrared spectrum detection method comprises the following steps:
acquiring infrared spectrum data of a sample and preprocessing the infrared spectrum data;
performing waveband selection based on the preprocessed sample infrared spectrum data to obtain the infrared spectrum data of the sample to be detected;
automatically establishing a near-infrared qualitative inspection model based on infrared spectrum data of a sample to be detected;
judging a near-infrared qualitative inspection model by adopting a multi-stage release strategy, and judging whether the near-infrared qualitative inspection model is correct or not;
if the result is correct, checking according to the near-infrared qualitative checking model, and storing the checking result into a material database;
if the error is detected, the sample is not stored in the material database, and the sample is subjected to chemical detection.
Further, the acquiring and preprocessing sample spectrum data comprises:
acquiring sample spectral data;
smoothing and denoising the sample spectrum data by adopting a convolution smoothing method;
eliminating the influence of surface property difference and particle size scale of the sample spectral data subjected to smoothing noise reduction by using a multivariate scattering correction method;
and obtaining the spectrum data of the preprocessed sample.
Further, the obtaining of the spectral data of the sample to be detected by performing band selection based on the preprocessed spectral data of the sample includes:
carrying out correlation calculation on the absorbance vector work corresponding to each wavelength in the correction set spectrum array and the concentration vector of the component to be measured in the concentration array;
the more information of the wavelength with the larger absolute value of the corresponding correlation coefficient, the larger correlation coefficient is selected as the waveband to be measured.
Further, the determining the near-infrared qualitative inspection model by using the multi-level release strategy to determine whether the near-infrared qualitative inspection model is correct includes:
starting a first-level release strategy, wherein the first-level release strategy is to perform correlation calculation on the near-infrared qualitative inspection model and the models in the spectrum model library, and if the correlation coefficient with the similar material models is above a first threshold value and the correlation coefficient with the different material models is below a second threshold value, judging that the correlation coefficient is correct; otherwise, starting a secondary release strategy;
the secondary release strategy is to establish an SIMCA model based on the infrared spectrum data of the sample to be detected, analyze the SIMCA model and the models in the spectrum model library through the similarity coefficient, and judge the accuracy if the correlation coefficient with the similar materials is above a second threshold value; otherwise, starting a three-level release strategy;
the three-level release strategy is to establish a support vector machine model based on infrared spectrum data of a sample to be detected, analyze the support vector machine model and a model in a spectrum model library through a similarity coefficient, and judge that the model is correct if a correlation coefficient with the similar material is above a second threshold value; otherwise, automatically judging the sample as the high-risk material.
According to some embodiments, the second aspect of the present invention provides a handheld near infrared spectrum detection device, which adopts the following technical solutions:
a handheld near infrared spectrum detection device comprises a handheld terminal and a PC terminal, wherein the handheld terminal is connected with the PC terminal through a Bluetooth module;
the handheld terminal comprises a shell, a miniature near-infrared spectrometer, a controller and a Bluetooth module, wherein the miniature near-infrared spectrometer, the controller and the Bluetooth module are integrated in the shell;
the controller is connected with the PC terminal through the Bluetooth module and sends sample spectrum data collected by the micro near-infrared spectrometer to the PC terminal;
the PC terminal analyzes the sample spectrum data, and the analysis process is as the steps in the hand-held near infrared spectrum detection method according to the first scheme.
Further, the housing is divided into a head part and a hand-held part; the head is divided into two compartments, and a fiber-optic probe is fixed in a first compartment; a micro near-infrared spectrometer is fixed in the second compartment; the hand-held part is divided into three compartments, a controller is fixed in the third compartment, and the controller is connected with the micro near-infrared spectrometer; a Bluetooth module is fixed in the fourth compartment and is connected with the controller; a power supply module is fixed in the fifth compartment;
the head part is provided with a spiral switch, and the bottom of the head part is provided with a buckling switch; the handheld part is provided with a power control button and an anti-slip ring sleeve.
Furthermore, the bottom of the handheld part is covered, and the front end of the head is provided with a threaded sealing cover for connecting an optical fiber probe, so that the near-infrared instrument is convenient to overhaul.
Furthermore, the thread sealing cover is provided with two layers of clamping grooves which are respectively used for connecting the disposable probe sleeve and the metal protective cover.
Furthermore, a disposable probe sleeve is sleeved on the optical fiber probe.
Furthermore, the joint of the front end threaded sealing cover and the head shell is also provided with a sealing ring matched with the front end threaded sealing cover.
Compared with the prior art, the invention has the beneficial effects that:
the invention carries out model judgment on the sample by a multi-stage modeling method, firstly adopts a correlation coefficient method to determine the correlation coefficient of the sample, and the sample can be released within a specified range to meet the model standard. If the judgment is not carried out within the specified range, the next-level model judgment is carried out, until the third-level judgment is finished, the model criterion cannot be met, the materials are automatically assigned as high risk, the software rejects the materials to be put in storage and suggests the traditional chemical method for further confirmation, the risk of releasing judgment errors by a single method is avoided, and the method is favorable for application and popularization in the aspect of detection of raw and auxiliary materials of medicines.
The handheld terminal has the advantages that the structure is simplified, the size is reduced, the carrying is lighter, the contact detection is realized through the exposed design of the optical fiber probe, and the accuracy of spectral information is improved; the modeling automation degree is high, the modeling precision can be greatly improved, the anti-interference capability of the model is stronger, the sampling and modeling speed can be improved, the human error is reduced, and the intelligent management of data is facilitated; the disposable probe sleeve prevents the pollution of raw and auxiliary materials in the acquisition process; the shock-proof arrangement of the shell can improve the stability and the service life of the instrument; the Bluetooth communication enables the spectrum detection freedom degree to be better; the multistage release discrimination strategy realizes rigor and reasonability of unknown sample judgment.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a flow chart of predictive model automatic optimization and predictive spectrum modeling of a multi-level release system according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a handheld NIR spectroscopy inspection and automated modeling system according to an embodiment of the present invention;
FIG. 3 is a functional block diagram of a hand-held NIR spectroscopy detection and automatic modeling system according to an embodiment of the invention;
FIG. 4 is an instrument parameter configuration page of the hand-held NIR spectroscopy detection and automated modeling system according to an embodiment of the present invention;
in the figure: 1-a fiber optic probe; 2-sealing the cover; 3-micro near-infrared spectrometer; 4-a shock-proof device; 5-a controller; 6-indicator light; 7-power key; 8-a Bluetooth module; 9-a battery; 10-bottom cover; 11-antiskid rubber sleeve; 12-a protective cover; 13-a housing.
Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. 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.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an", and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiments and features of the embodiments of the present invention may be combined with each other without conflict.
Example one
As shown in fig. 1 to fig. 4, the present embodiment provides a handheld near infrared spectrum detection method, which includes the following steps:
a handheld near infrared spectrum detection method comprises the following steps:
acquiring infrared spectrum data of a sample and preprocessing the infrared spectrum data;
performing waveband selection based on the preprocessed sample infrared spectrum data to obtain the infrared spectrum data of the sample to be detected;
automatically establishing a near-infrared qualitative inspection model based on infrared spectrum data of a sample to be detected;
judging a near-infrared qualitative inspection model by adopting a multi-stage release strategy, and judging whether the near-infrared qualitative inspection model is correct or not;
if the result is correct, checking according to the near-infrared qualitative checking model, and storing the checking result into a material database;
if the error is found, the sample is not stored in the material database, and the sample is subjected to chemical detection.
Specifically, the acquiring and preprocessing sample spectral data includes:
acquiring sample spectral data;
smoothing and denoising the sample spectrum data by adopting a convolution smoothing method;
eliminating the influence of surface property difference and particle size scale of the sample spectral data subjected to smooth noise reduction by using a multivariate scattering correction method;
and obtaining the spectrum data of the pretreated sample.
Specifically, the obtaining of the spectral data of the sample to be detected by performing band selection based on the preprocessed spectral data of the sample includes:
carrying out correlation calculation on the absorbance vector work corresponding to each wavelength in the correction set spectrum array and the concentration vector of the component to be measured in the concentration array;
the more information of the wavelength with the larger absolute value of the corresponding correlation coefficient, the larger correlation coefficient is selected as the waveband to be measured.
Specifically, the determining, by using a multi-stage release strategy, a near-infrared qualitative inspection model to determine whether the near-infrared qualitative inspection model is correct includes:
starting a first-level release strategy, wherein the first-level release strategy is to perform correlation calculation on the near-infrared qualitative inspection model and the models in the spectrum model library, and if the correlation coefficient with the similar material models is above a first threshold value and the correlation coefficient with the different material models is below a second threshold value, judging that the correlation coefficient is correct; otherwise, starting a secondary release strategy;
the secondary release strategy is to establish an SIMCA model based on the infrared spectrum data of the sample to be detected, analyze the SIMCA model and the models in the spectrum model library through the similarity coefficient, and judge the accuracy if the correlation coefficient with the similar materials is above a second threshold value; otherwise, starting a three-level release strategy;
the three-level release strategy is to establish a support vector machine model based on infrared spectrum data of a sample to be detected, analyze the support vector machine model and a model in a spectrum model library through a similarity coefficient, and judge that the model is correct if a correlation coefficient with the similar material is above a second threshold value; otherwise, automatically judging the sample as the high-risk material.
The invention also discloses an automatic optimization and multistage release processing system for the prediction model. The software subsystem design comprises an upper computer software subsystem design and a lower computer software subsystem design. The upper computer software is Windows 10 system software and is connected with the controller and the PC end through the Bluetooth module. The controller adopts STM32 singlechip, and the lower computer adopts uKeil5 development environment to develop, directly downloads into STM32 singlechip after compiling and can use.
In the embodiment, the software is mainly divided into five layers, wherein the bottom layer is a communication layer for receiving data and sending instructions; the middle includes an infrastructure layer to provide the system with a processor and memory; the spectrum algorithm layer comprises two modules, wherein one time bottom layer processing module is used for storing a preprocessing method, a qualitative analysis algorithm and an optimization algorithm; a time top control module realizes the processing and calling of the algorithm; the spectrum model layer is used for searching the existing spectrum model, storing a new spectrum model and converting the spectrum model. And the human-computer interaction layer at the top layer is used for interface display. A software system architecture block diagram is shown in figure 1.
In this example, data are transmitted through bluetooth module at PC end and device end, have realized sample data detection's convenient, flexibility. The communication method of the Bluetooth module comprises the following steps:
the Bluetooth module is connected: the operation of the program is performed according to the sequence of opening the Bluetooth module, initializing the Bluetooth module, inquiring the nearby Bluetooth module, establishing connection, sending data/files, disconnecting the connection and closing the Bluetooth module, and is triggered by a button of the operation area. After the initialization of the Bluetooth module is completed, the local Bluetooth module address is displayed in the information area, and after the inquiry is completed, the inquired remote Bluetooth module address is displayed in the information area, and then the connection establishment can be initiated. The local PC end Bluetooth module is used as a main device, firstly, the inquiry is initiatively initiated, after the device Bluetooth module device is inquired, the connection is initiatively initiated, and after the connection is established, the data can be transmitted between the PC end Bluetooth module and the device Bluetooth module.
Data transmission: after the connection of the Bluetooth module between the singlechip end and the PC end is established, data can be transmitted between the singlechip end and the PC end. The PC end sends out an instruction, and the singlechip end controls the near infrared to carry out spectrum acquisition together after receiving the data. When the single chip computer is used as the slave device, if the automatic return opening is set, the received data can be returned to the PC. The process of file transmission between the single chip computer end and the PC end is the process of equipment interaction at the two ends. The single chip microcomputer needs an external memory to temporarily store data to be transmitted, so that a simple FAT16 file system is realized by taking the MMC card as the external memory and programming the MMC card on the single chip microcomputer, so that the single chip microcomputer receives a file from a PC end through a Bluetooth module wireless interface and stores the file in a FAT16 format matched with the PC, and finally, the aim is that the received file can be identified by connecting the MMC card to the PC through a card reader after the MMC card is taken out from the single chip microcomputer end. The files transmitted from the PC are all in a certain format, which is commonly the files in the formats of FAT16 and FAT32, and the files received from the PC must be stored in a format matching with the files before the files are successfully transmitted. The process is as follows:
1. the single chip microcomputer firstly sends file sending request information to the PC end, and the PC returns file sending permission information to the single chip microcomputer after receiving the request.
2. And after receiving the information which is allowed to be transmitted, the single chip microcomputer opens the file to be transmitted in the MMC, acquires file information and file data, transmits the file information to the PC end, creates an empty file according to the file information after receiving the file information by the PC end, and returns file creation completion information to the single chip microcomputer after the file creation completion.
3. After the single chip microcomputer learns that the PC file is established, the file data can be sent, and the file data is sent in a segmented sending mode.
4. After the data transmission of the single chip microcomputer is finished, the single chip microcomputer sends file transmission finishing information to the PC, and the PC closes the current file.
In this example, the upper computer software is Windows 10 system software, and is connected to the analyzer through the bluetooth module, and the software requirements are as follows:
(1) the serial port has the functions of: the serial port can be automatically scanned, the available serial port can be displayed on software, the baud rate of the serial port can be set, and a connection button and a disconnection button are arranged.
(2) A system switch: and a system switch function is set, other functions cannot be used when the system is closed, and other functions can be used only when the system is opened.
(3) NIR spectral modeling optimization function: preprocessing the received data, calling a pattern recognition algorithm, fixing a primary modeling method as a correlation coefficient method, and selecting more than two levels according to the attributes of the sample.
(4) NIR spectral qualitative model storage function: and storing the established spectrum model to a specified path, and calling and checking at any time.
(5) The tungsten lamp has the following control functions: switch capable of setting tungsten lamp
(6) PID parameter setting function: PID parameters, such as KP, KI, KD, can be set.
(7) NIR spectroscopy shows the function: and displaying the acquired spectrum in real time, wherein the horizontal axis is time, and the vertical axis is the corresponding current value.
The software is written in Python 3.7 language and Pycharm development environment. Writing upper computer control software according to software requirements and a software system structure diagram, and dividing the software into four functional modules: the device comprises a user interaction module, an instrument parameter configuration module, a spectrum acquisition module, a spectrum modeling module and a model storage module.
The user interaction module is an interface module for a user to use the software, so that the man-machine interaction is more friendly. The module was developed using PyQt5, in which two tabs were provided to support the configuration of instrument parameters and the acquisition of spectra, respectively. All the functions of keys, tabs and graphical displays are also performed in this module.
The instrument parameter configuration module mainly uses a Pyserial module to complete the bottom layer communication function, and the Pyserial module encapsulates the access to a serial port (serial port). It provides the backend for Python running on Windows, OSX, Linux, BSD and IronPython. The module name "serial" will automatically select the appropriate backend. The instrument parameter configuration module completes the main functions of software, such as traversing all ports of the upper computer, determining the model of the connected detector, establishing long connection between the upper computer and the detector NIRONE 1.7, sending user instructions and receiving data of the NIRONE 1.7.
After the spectrum acquisition module establishes connection between an upper computer and a Niron Sensor by Pyserial and configures detection parameters, the spectrum acquisition module can be responsible for acquisition of a dark current spectrum through a 'dark current' key, a 'reference spectrum' key is responsible for acquisition of a reference spectrum, a 'measure' key is responsible for acquisition of a sample spectrum, and after acquisition of the dark current and the reference spectrum is completed, a sample spectrum display image can be automatically converted into an absorbance spectrum according to a formula when the sample spectrum is acquired, wherein the conversion formula is as follows:
a spectrum modeling module: the method is divided into two modes, namely an automatic modeling mode, namely a method for automatically modeling by using an optimal algorithm pushed by a machine, and the other manual modeling mode, namely a method for manually selecting the needed algorithm for modeling stored in the machine according to the attribute of a measured sample by an operator, and recommending and realizing one-key modeling by forbidding to call the optimization algorithm. The module is mainly completed by a NumPy, SciPy (Scientific Python) and Matplotlib module.
The model storage module is mainly responsible for storing data during data acquisition, software can automatically store the data when a user acquires the data, and due to the characteristic that the xlsxwriter module can store the data for many times, the module is used for completing the development of an automatic storage function, but the module supports the writing of 256 columns at most and cannot meet the requirement of storing the data when the resolution ratio is 1. To solve this problem, the resolution is determined before reading and writing data, and if the number of data columns exceeds 256 rows, the data columns are automatically divided into two tables (sheets) for storage. Besides the system automatically saves data, the system supports manual saving of users, and the manual saving function is realized by the development of an xlwt module.
In this example, opening the system main interface may occur: parameter configuration, file storage, spectral measurement, modeling and verification, sample determination, viewing and the like. After the system is operated, the configuration menu can start to work. The modeling process is as follows:
(1) to the spectral parameter configuration interface. The instrument parameter configuration page is entered, as shown in fig. 4, and the port is selected on the page, i.e., the type and number of near-infrared instruments to be used are selected. Selecting proper instrument detection parameters, wherein the average wavelength is wavelength average, the average scanning is scan average, and the average wavelength and the average scanning are parameters for determining the frequency of spectrum scanning; the resolution is a parameter for setting the number of wavelength points, and when the resolution is 1, an energy value is collected every 1 nm; the scanning interval is the time interval of spectrum acquisition in the automatic scanning mode, and if 5s is set, the instrument automatically acquires a spectrum every 5 seconds. If the basic parameters are not needed to be set, a 'default value' button can be clicked, the software can automatically fill in the conventional basic parameters, and in the default state, the average wavelength is 100, the average scanning is 1, the resolution is 1, and the scanning interval is 5 s.
(2) And selecting a new file path from a file menu and setting a file name. Selecting a collected spectrum data storage path, and directly inputting a path to be stored and a file name in a label column under 'project name modification'; if the path position is selected, a 'save position' button can be clicked, and a popup window is popped up to select the save position and the name of the file to be saved.
(3) And receiving spectral data, and displaying the original spectrogram of the sample in real time. Clicking a start configuration button, transmitting detection transmission required by the instrument to a Nirone Sensor through a Bluetooth module, and displaying the configuration completed!by a software command feedback column! And switching to a spectrum measurement interface at the moment, collecting the spectrum and displaying the spectrum collection interface in real time.
(4) And selecting automatic spectrum modeling and manual modeling. And clicking a modeling and verifying button to provide two options of automatic modeling and manual modeling, selecting an automatic modeling option system to automatically model, and displaying a model calibration and verification result after the model is built or is verified. If satisfied with the model validation results, information such as model name, sample type, etc. may then be entered in the dialog box and saved by clicking the "save to model library" button. If the manual modeling button is selected, a manual modeling algorithm selection interface is popped up, the selection of methods such as preprocessing methods and mode recognition methods is included, the selection is performed from algorithms stored in a spectrum algorithm layer of software, the 'modeling' button is pressed, the system calls corresponding algorithms in the software according to the method selected by a user to process spectrum data, corresponding results are displayed after the processing is completed, and the storage process and the automatic modeling are performed. The function realizes the high efficiency and rapidness of automatic modeling, and also meets the accuracy and the science of manual modeling.
(5) The system uses the sample determination menu to complete the qualitative judgment of the unknown sample based on the model base. First, a sample measurement menu is clicked, and a sample measurement dialog box is opened. And clicking a sample file reading button, and selecting sample optical latent data to be initially determined from an opened dialog box. And finally, selecting a prediction model to be used from the existing model list, and clicking a measurement button after the selection to realize measurement.
The automatic modeling of raw and auxiliary materials and the automatic multi-stage release method of unknown samples in the embodiment are as follows:
selecting a spectrum file to be modeled, and dividing the spectrum of the sample into a calibration set sample and a verification set sample. And the calibration set sample is used for spectral modeling, and the verification set sample is used for verifying the performance of a back model. The samples in the calibration set should contain all chemical constituents that may be present in future samples to be tested; the concentration (or property) range should exceed what may be encountered in future samples to be tested.
And preprocessing the spectral data of the sample. Modern measuring instruments are more precise, and especially in the original spectrum acquired by the near infrared spectrum, besides sample composition information, the performance of the established model is influenced by the amount of noise. In order to eliminate the influence of noise or other external factors on the spectrum, the model needs to be preprocessed to improve the accuracy of model prediction. The software is embedded with a Savitsky-Golay convolution smoothing method and a Multivariate Scattering Correction (MSC), wherein the Savitsky-Golay convolution smoothing method is used for denoising and smoothing, and the MSC is selected to eliminate the influence caused by surface property difference and particle size of samples because most of raw and auxiliary materials are solid particles.
(1) The Savitsky-Golay convolution smoothing method assumes that noise contained in a spectrum is zero-mean random white noise, performs weighted average processing on data through multiple measurements, reduces the noise and improves the signal-to-noise ratio, and the average value at the wavelength k after smoothing is as follows:
in the formula, h i For smoothing coefficients, H is a normalization factor, H ═ H i Each measurement is multiplied by a smoothing factor h i With the aim of minimizing the effect of smoothing on the useful information, h i Can be solved by polynomial fitting based on the principle of least squares.
(2) The Multivariate Scatter Correction (MSC) is applied to a spectrum x (1 × m), and the specific algorithm of the MSC is as follows:
A. meterCalculating the average spectrum of the calibration set samples
(i.e., the "ideal spectrum");
B. x is reacted with
The linear regression is carried out, and the linear regression is carried out,
solving for b by least squares 0 And b;
C.x MSC ＝(x-b 0 )/b 0 ；
and selecting the wave bands of the spectrum subjected to denoising and smoothing. When near-infrared correction modeling is carried out, the calculation amount is greatly increased by adopting full-spectrum calculation, and the correlation between key spectrum information and index property pieces is influenced by the existence of certain wave band spectrum information, so that wave band selection needs to be carried out on a spectrum, and a wave band with strong correlation on measured properties is found out. The invention adopts a common method correlation coefficient method of wave band selection. And (3) carrying out correlation calculation on the absorbance value corresponding to each wavelength in the correction set spectrum array and the concentration value to be detected, wherein the larger the correlation coefficient is, the more the information content of the wavelength is.
The correlation coefficient method is to perform correlation calculation on the absorbance vector work corresponding to each wavelength in the correction set spectrum array and the concentration vector y of the component to be measured in the concentration array, and the information of the wavelength with the larger absolute value (or decision coefficient) of the corresponding correlation coefficient is more. The correlation coefficient R is calculated by the following formula:
wherein the content of the first and second substances,
n is the number of samples corrected.
And (3) performing automatic optimization modeling on the spectrum subjected to pretreatment and waveband selection, and adopting a first derivative method, SG smoothing method and SNV method. The qualitative modeling method comprises qualitative discrimination modeling, cluster modeling and the like, the invention adopts a cluster analysis method to carry out spectrum modeling, and important components of the cluster analysis are the distance between samples, the distance between classes, the merging mode and the cluster number. The method comprises the following steps of adopting an automatic multi-level release strategy, namely modeling a correction set and a verification set of a sample spectrum by a first-level discrimination method correlation coefficient method, a second-level discrimination method SMICA method and a third-level discrimination method SVM method in sequence, storing a model in a spectrum model library, and verifying the model by the verification set into four conditions:
the first-level discrimination method is a similarity coefficient method, and after discrimination, the correlation coefficient of the verification set spectrum and the calibration set spectrum with the material spectrum is required to be above 0.97, and the correlation coefficient with different material spectra is required to be below 0.85; materials meeting the condition of the correlation coefficient method are successfully distinguished, and the materials are a first-level release strategy;
when the correlation coefficient between the verification set spectrum and the correction set and the material is below 0.97 and above 0.85, the physical properties or batch difference of the verification set spectrum and the correction set may exist, and the system automatically starts a secondary release strategy, namely, a SIMCA (soft independent modeling of class analysis) model is established for further attributing the material;
when the correlation coefficient between the verification set spectrum and the calibration set and the materials is below 0.85, the system automatically starts a second-level release strategy and a third-level release strategy, namely, the system further belongs to the material type through a Support Vector Machine (SVM) model;
when the materials cannot be judged correctly by the three-level release strategy, the materials are automatically classified as high risk, and software rejects the materials to be put in storage and suggests a traditional chemical method for further confirmation.
The similarity coefficient method considers each sample as a point of m-dimensional space (m variables) in which the degree of affinity and sparseness between samples is defined.
Wherein
The average values of all the characteristic variables of the ith and jth samples respectively. The closer the two samples are, the closer the coefficient of similarity between them is to 1 (or-1).
In the method, the first-level release method is a correlation coefficient method, and the second-level and third-level release methods can be selected as different qualitative discrimination methods according to the properties of the sample, such as an artificial neural network method (ANN), a K nearest neighbor method (KNM) and the like.
The automatic prediction model establishing and optimizing processing system automatically establishes and optimizes a prediction model for a sample to be tested through five functional modules, namely a user interaction module, an instrument parameter configuration module, a spectrum acquisition module, a spectrum modeling module and a model storage module. The problem that the traditional modeling method needs a large amount of manpower and material resources to collect spectrums of various samples in the sample collection process, time and labor are consumed, and operation errors caused by personnel are generated, so that model precision loss is caused. The automatic spectrum modeling does not need excessive intervention of personnel, professional modeling and judgment of qualified parameter conditions, can improve the modeling speed, reduce human errors and facilitate intelligent data management.
Example two
As shown in fig. 2, the present embodiment provides a handheld near infrared spectrum detection apparatus, including a handheld terminal and a PC terminal, where the handheld terminal is connected to the PC terminal through a bluetooth module;
the handheld terminal comprises a shell, a miniature near-infrared spectrometer, a controller and a Bluetooth module, wherein the miniature near-infrared spectrometer, the controller and the Bluetooth module are integrated in the shell;
the controller is connected with the PC terminal through the Bluetooth module and sends sample spectrum data collected by the micro near-infrared spectrometer to the PC terminal;
the PC terminal analyzes the sample spectral data, and the analysis process is as the step in the hand-held near infrared spectrum detection method of any one of claims 1 to 4.
Wherein the shell is divided into a head part and a handheld part; the head is divided into two compartments, and a fiber-optic probe is fixed in a first compartment; a micro near-infrared spectrometer is fixed in the second compartment; the hand-held part is divided into three compartments, a controller is fixed in the third compartment, and the controller is connected with the micro near-infrared spectrometer; a Bluetooth module is fixed in the fourth compartment and is connected with the controller; a power supply module is fixed in the fifth compartment;
the head part is provided with a spiral switch, and the bottom of the head part is provided with a buckling switch; the handheld part is provided with a power control button and an anti-slip ring sleeve.
The bottom sealing cover of the handheld part is arranged, the front end of the head part is provided with a threaded sealing cover for connecting an optical fiber probe, and the near-infrared instrument is convenient to overhaul.
The thread sealing cover is provided with two layers of clamping grooves which are respectively used for connecting the disposable probe sleeve and the metal protective cover.
And the optical fiber probe is also sleeved with a disposable probe sleeve.
The joint of the front end thread sealing cover and the head shell is also provided with a sealing ring matched with the front end thread sealing cover.
This implementation provides a hand-held type near infrared spectroscopy detects and automatic modeling system to the halogen tungsten lamp is the light source, shines in direct contact's sample after the collimation, through the transmission after, guarantees the parallel by optic fibre receipt of light path through one-level collimation again, and the light of this moment is the spectrum that has contained a large amount of sample information, and optic fibre transmits the spectrum to the spectrum appearance. The spectrum enters a spectrometer and is subjected to light splitting through a Fabry-Perot interferometer based on MEMS, a large amount of stray light is filtered, and finally, the indium gallium arsenic photoelectric detector converts the split spectrum into an electric signal and stores the electric signal in a controller. The controller transmits the sample spectrum information to the PC end through the Bluetooth module. After receiving the spectrum data, the PC end carries out format conversion on the spectrum data, automatically stores the spectrum data to a set storage path, and forms a spectrum image through the man-machine interaction module.
After the sample data is collected, the software system selects the collected data to process, selects a preprocessing method, eliminates the influence of various factors on the spectral data and selects a proper waveband to analyze and process; selecting a proper mode identification method, namely a qualitative modeling method, from the preprocessed spectral data, performing multi-method multi-stage modeling on the sample, wherein the first-stage modeling method is a correlation coefficient method, and the modeling above the second-stage modeling can be selected to be different qualitative discrimination methods according to the properties of the sample, such as an artificial neural network method (ANN), a K nearest neighbor method (KNM) and the like; and finally, distinguishing the unknown sample by a multi-stage release method. Determining a mode identification method according to the type of a sample to be detected, and calculating spectral distance to screen out-of-range sample data; after the automatic modeling process is completed, the system displays the results of all the steps on a PC (personal computer) end through a human-computer interaction module, and an operator judges whether the results are stored or not according to the results. If the result is stored, the corresponding sample information and the model are stored in a storage module of the PC terminal together, so that subsequent calling and checking are facilitated. The unknown sample detection can search out the established model according to the sample attribute, and bring the spectral data into the model to carry out qualitative judgment on the unknown sample. Aiming at the modeling of a sample with special requirements, the system provides a manual modeling mode, manually selects an algorithm required by modeling, and realizes personalized modeling. The method realizes the rapidness and high efficiency of modeling, and is suitable for one-line work of modeling a large number of products. And (4) displaying a data bar when each step is completed, and popping up a warning window by the system if an error occurs in the midway.
In this example, the main component types mainly include: halogen tungsten lamp as light source (model: 997418-21), micro NIR spectrometer (NIRONE 1.7), controller (STM32), Bluetooth module (4.0BLE)
The invention discloses a handheld near infrared spectrum detection and automatic modeling system subsystem and a handheld near infrared spectrum detection hardware system, which have the overall requirements of compact structure, small volume, higher portability and higher shock resistance. In this example, the hardware subsystem functional structure includes: (1) collecting spectrometer data and transmitting the data to a computer host through a serial communication interface; (2) and receiving a computer instruction and executing the light source switch.
This hand-held type near infrared spectroscopy detects hardware system includes: in this example, the housing; the micro near-infrared spectrometer, the Bluetooth module, the power module and the controller are integrated in the shell; an optical fiber probe and a protective cover; a disposable probe sleeve. The optical fiber probe is connected with a near-infrared spectrometer inside the shell; the controller is electrically connected to the infrared spectrometer; the Bluetooth module is electrically connected with the controller; the power supply module is electrically connected with the controller; the disposable probe sleeve is sleeved on the optical fiber probe during spectrum acquisition. In the example, the hardware subsystem uses the STM32 main control chip as a control center, the main control chip and the computer host communicate through a serial communication port, the serial port is short for serial port, the hardware subsystem is characterized in that the communication circuit is simple, and bidirectional communication can be realized only through a Bluetooth module. The main control chip can receive instructions sent by the computer host, such as spectrometer instructions and light source switching instructions. Meanwhile, the main control chip can report the original data collected by the spectrometer to the computer host.
In this example, the housing is provided as a handheld structure and is divided into two parts: a head portion and a hand-held portion. A housing; the optical fiber probe protective cover, the thread sealing cover and the micro near-infrared spectrometer with the optical fiber probe are sequentially connected, wherein the optical fiber probe protective cover is contained in the head; the controller, bluetooth module, power and bottom closing cap that handheld portion accomodate. The shell is provided with a switch button arranged on the handheld part; the key switch is used for starting and closing the power supply of the hardware device. The two ends of the shell are provided with openable sealing covers, and the spiral sealing cover at the head is used for disassembling the optical fiber probe and is convenient for maintaining internal instruments; two layers of clamping grooves are formed in the edge of the sealing cover, the front layer of clamping groove is used for mounting the disposable probe sleeve, and the second layer of clamping groove is used for mounting the protective cover; the bottom sealing cover is used for disassembling the battery, adopts a buckling sealing cover mode, is arranged on the USB connecting port and is used for charging the battery. The handheld part of the shell is sleeved with an anti-slip ring sleeve to prevent the instrument from falling off in the operation process; the inside gasbag that is equipped with of casing takes precautions against earthquakes and sets up stable near-infrared spectrum appearance, including buffering gasbag and protection pad double-deck protection, closely laminates near-infrared spectrum appearance. Portable near-infrared device internal component is mostly precision instrument, and small vibrations all have certain influence to the testing result, and getting of instrument in the operation process also can lead to internal component's connection to go wrong, and the shockproof setting of this kind of gasbag can be fine the protection instrument avoid receiving outside influence. Each internal element is divided by a corresponding partition plate, and the internal part is divided into five small compartments which are respectively a near-infrared spectrometer, an optical circuit board, a controller, a Bluetooth module and a battery.
In this embodiment, the housing includes: the sealing cover, the front end screw sealing cover, the main body, the bottom cover and the indicating lamp on the side face are connected in sequence. The bottom cover is connected with the hand-held part in a buckling mode, and the front end threaded sealing cover is connected with the main body part in a screwing mode. The head part of the device is used for accommodating and fixing the micro near-infrared spectrometer and the mounting structure of the circuit board. The main part is handheld form, and its detachable is two baffles along axial direction, and two arc combination lines become the cavity that is used for holding controller, bluetooth module and battery. The bottom cover is used for sealing the bottom of the hand-held part body. For another example, a sealing ring matched with the spiral sealing cover is arranged at one end of the shell, and the sealing ring is used for improving the sealing performance of the sealing cover when the sealing cover is closed, protecting a probe of the micro near-infrared spectrometer and avoiding the interference of ambient light on spectrum collection. For another example, the housing is provided with a power button electrically connected to the power control module. The operator starts supplying power to each electronic component and starts the apparatus by pressing the power key.
In this example, the probe end of the near infrared spectrometer is disposed toward the spiral cover. The preferred near infrared spectrometer is a high performance, compact, reliable near infrared spectroscopy sensor. For example, in this embodiment, the near infrared spectrometer is a NIRONE 1.7 micro NIR spectrometer, and the NIR spectrometer sensor can be controlled by PC through a sensor control software with friendly interface. The replacement of the spectrometer does not cause the linkage of other parts, namely 4 wires (a power supply positive wire, a power supply negative wire, a serial port receiving end and a serial port transmitting end) led out from the existing spectrometer are connected to the XH2.54 female head through transforming the spectrometer and are in butt joint with the XH2.54 male head led out from the controller, so that the replacement of the spectrometer is realized.
In this example, the bluetooth module is electrically connected to the controller. The Bluetooth communication module is used as a current universal short-distance wireless communication module and can be conveniently in wireless connection with a PC terminal. As shown in fig. 2, in this embodiment, a preferred bluetooth module 4.0BLE is taken as an example, and it may be connected to a thousand controllers, and implement data transmission with a PC terminal by means of bluetooth module transmission.
In this example, the power supply includes: the battery is used as a built-in movable storage component, and the rechargeable battery is used as a device power supply. In order to realize green, environmental protection and convenience, the power supply element of the hardware subsystem, namely the power supply module, selects the rechargeable battery which can be recycled, and the power supply module can realize the purpose of obtaining electric energy by connecting an external power supply from a power supply interface. For example, the power source may be a rechargeable battery housed in the housing, and the battery may be charged through the power source interface. A lithium-ion electrochemical cell is preferred, which has a graphitic anode material or a lithium titanate anode material, and a lithium iron phosphate cathode material suitable for rapid recharging of rechargeable batteries, and a battery case for holding the battery is also provided.
In this example, the disposable probe cover is used for detecting the optical fiber probe of different samples. As shown in figure 2, the shape and the size are manufactured according to the relevant information of the optical fiber probe, when the detection is carried out, the protective cover of the probe is removed, the disposable probe sleeve is arranged on the optical fiber probe, the disposable probe sleeve and the optical fiber probe are tightly attached, and the disposable probe sleeve and the optical fiber probe are spliced with the first clamping groove at the front end of the shell to finish the detection. In the embodiment, the disposable probe sleeve is preferably made of PMMA plastic, so that the optical transmittance is high, and the influence on the detection result is avoided.
In this example, the controller is electrically connected to the micro near-infrared spectrometer. The controller is used as a driving circuit of the micro near-infrared spectrometer and is used as a control module. In this embodiment, a NanoPi-Neo circuit board can be used to build the driving circuit board. The controller is used as a control center of the hardware subsystem; the controller STM32 and the upper computer communicate through a serial communication port, and bidirectional communication can be realized only by the Bluetooth module. The controller can receive instructions sent by an upper computer, such as a spectrometer instruction and a light source switching instruction. Meanwhile, the controller can also report the original data collected by the spectrometer to the upper computer.
An STM32 main control chip is used as the main control chip, and the specific model is STM32F103C8T 6.
The handheld near infrared spectrum detection hardware system is used as a handheld terminal during sample detection, and has the function of data acquisition of near infrared spectrum detection. The miniature near-infrared spectrometer is a small-volume spectrometer, can compress the occupied space of the whole equipment, and is favorable for the miniaturization improvement of the equipment.
The micro near-infrared spectrometer is accommodated in the shell, and the spectrum collection of the micro near-infrared spectrometer is controlled through a collection button at the handheld part of the shell.
The controller is used for driving the micro near-infrared spectrometer to operate, and the wireless communication module is used for realizing data exchange with an external PC terminal, so that the acquisition parameters of the micro near-infrared spectrometer are set, and the data analysis and processing are carried out by utilizing the PC terminal. The rechargeable battery is used as a power supply unit, so that green, environment-friendly and efficient energy utilization is realized.
In addition, each electronic component can be fixed in the shell, and the working stability and the safety of each electronic component can be ensured.
Through the design, adopt miniature near-infrared spectrometer to carry out the spectral detection to utilize wireless communication module to set up the parameter acquisition who detects and data analysis processing process from handheld end separation to outside PC terminal, simplify the structure at hand-held type terminal, the volume reduces, carries more lightweight, is favorable to the application in the aspect of medicine raw and auxiliary materials detects and promotes.
As shown in fig. 2, the near infrared spectrum detection system includes the handheld near infrared spectrum detection device of any of the above embodiments, and further includes a PC terminal connected to the handheld near infrared spectrum detection device.
In summary, the embodiment of the invention provides a dedicated analyzer for detecting a spectrum of a pharmaceutical raw material and auxiliary material based on a NIRONE micro near infrared spectrometer, starting from a hardware subsystem and a software subsystem, wherein the analyzer is externally connected with a 220V power supply, and the voltage is converted to the voltage required by the operation of a main control chip and other components through a switching power supply. And the software system develops software of an upper computer and a lower computer, wherein the upper computer software system designs each functional interface, and each optimized chemometrics method is embedded into the software system to realize automatic establishment and optimization of the prediction spectrum detection model.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. A hand-held near infrared spectrum detection method is characterized by comprising the following steps:
acquiring infrared spectrum data of a sample and preprocessing the infrared spectrum data;
performing waveband selection based on the preprocessed sample infrared spectrum data to obtain the infrared spectrum data of the sample to be detected;
automatically establishing a near-infrared qualitative inspection model based on infrared spectrum data of a sample to be detected;
judging a near-infrared qualitative inspection model by adopting a multi-stage release strategy, and judging whether the near-infrared qualitative inspection model is correct or not;
if the result is correct, checking according to the near-infrared qualitative checking model, and storing the checking result into a material database;
if the error is found, the sample is not stored in the material database, and the sample is subjected to chemical detection.
2. The method of claim 1, wherein the acquiring and preprocessing sample spectral data comprises:
acquiring sample spectral data;
smoothing and denoising the sample spectrum data by adopting a convolution smoothing method;
eliminating the influence of surface property difference and particle size scale of the sample spectral data subjected to smooth noise reduction by using a multivariate scattering correction method;
and obtaining the spectrum data of the pretreated sample.
3. The method according to claim 1, wherein the obtaining the spectral data of the sample to be detected by performing band selection based on the preprocessed spectral data of the sample comprises:
carrying out correlation calculation on the absorbance vector work corresponding to each wavelength in the correction set spectrum array and the concentration vector of the component to be measured in the concentration array;
the more information of the wavelength with the larger absolute value of the corresponding correlation coefficient, the larger correlation coefficient is selected as the waveband to be measured.
4. The method according to claim 1, wherein the determining the near-infrared qualitative inspection model by using the multi-stage release strategy to determine whether the near-infrared qualitative inspection model is correct comprises:
starting a first-level release strategy, wherein the first-level release strategy is to perform correlation calculation on the near-infrared qualitative inspection model and the models in the spectrum model library, and if the correlation coefficient with the similar material models is above a first threshold value and the correlation coefficient with the different material models is below a second threshold value, judging that the correlation coefficient is correct; otherwise, starting a secondary release strategy;
the secondary release strategy is to establish an SIMCA model based on infrared spectrum data of a sample to be detected, analyze the SIMCA model and a model in a spectrum model library through a similarity coefficient, and judge the accuracy if a correlation coefficient with a similar material is above a second threshold value; otherwise, starting a three-level release strategy;
the three-level release strategy is to establish a support vector machine model based on infrared spectrum data of a sample to be detected, analyze the support vector machine model and a model in a spectrum model library through a similarity coefficient, and judge that the model is correct if a correlation coefficient with the similar material is above a second threshold value; otherwise, automatically judging the sample as the high-risk material.
5. A handheld near infrared spectrum detection device is characterized by comprising a handheld terminal and a PC terminal, wherein the handheld terminal is connected with the PC terminal through a Bluetooth module;
the handheld terminal comprises a shell, a miniature near-infrared spectrometer, a controller and a Bluetooth module, wherein the miniature near-infrared spectrometer, the controller and the Bluetooth module are integrated in the shell;
the controller is connected with the PC terminal through the Bluetooth module and sends sample spectrum data collected by the micro near-infrared spectrometer to the PC terminal;
the PC terminal analyzes the sample spectral data, and the analysis process is as the step in the hand-held near infrared spectrum detection method of any one of claims 1 to 4.
6. The apparatus according to claim 4, wherein said housing is divided into a head portion and a hand portion; the head is divided into two compartments, and a fiber-optic probe is fixed in a first compartment; a micro near-infrared spectrometer is fixed in the second compartment; the hand-held part is divided into three compartments, a controller is fixed in the third compartment, and the controller is connected with the micro near-infrared spectrometer; a Bluetooth module is fixed in the fourth compartment and is connected with the controller; a power supply module is fixed in the fifth compartment;
the head part is provided with a spiral switch, and the bottom of the head part is provided with a buckling switch; the handheld part is provided with a power control button and an anti-slip ring sleeve.
7. The hand-held near infrared spectrum detection device of claim 5, wherein a bottom sealing cover of the hand-held part is arranged, and a threaded sealing cover is arranged at the front end of the head part for connecting the optical fiber probe, so that the near infrared instrument is convenient to overhaul.
8. The hand-held near infrared spectroscopy apparatus of claim 6 wherein the screw cap has two layers of slots for connecting a disposable probe cover and a metal shield, respectively.
9. The hand-held near infrared spectroscopy apparatus of claim 5 wherein the fiber optic probe is further sleeved with a disposable probe cover.
10. The hand-held near infrared spectroscopy apparatus of claim 7 wherein the connection of the front threaded closure and the head housing further comprises a seal ring mating with the front threaded closure.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210842698.6A CN115128036A (en) | 2022-07-18 | 2022-07-18 | Handheld near infrared spectrum detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210842698.6A CN115128036A (en) | 2022-07-18 | 2022-07-18 | Handheld near infrared spectrum detection method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115128036A true CN115128036A (en) | 2022-09-30 |
Family
ID=83383356
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210842698.6A Pending CN115128036A (en) | 2022-07-18 | 2022-07-18 | Handheld near infrared spectrum detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115128036A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116165165A (en) * | 2023-04-25 | 2023-05-26 | 四川威斯派克科技有限公司 | Detection method for online real-time release of raw and auxiliary materials of medicines |
| CN116660206A (en) * | 2023-05-31 | 2023-08-29 | 浙江省农业科学院 | Crop yield estimation method and system |
| CN116660185A (en) * | 2023-06-02 | 2023-08-29 | 成都信息工程大学 | Multi-wavelength heavy metal ion solution detection system and detection method thereof |
| CN116953488A (en) * | 2023-09-19 | 2023-10-27 | 深圳市东陆科技有限公司 | Monitoring method for integrated photoelectric chip |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104237060A (en) * | 2014-10-05 | 2014-12-24 | 浙江大学 | Multi-index quick detection method of honeysuckle |
| CN104792652A (en) * | 2015-05-02 | 2015-07-22 | 浙江大学 | Multi-index rapid detection method for radix astragali |
| CN104833651A (en) * | 2015-04-15 | 2015-08-12 | 浙江大学 | Honeysuckle concentration process online real-time discharging detection method |
| CN106226264A (en) * | 2016-05-05 | 2016-12-14 | 江苏康缘药业股份有限公司 | Herba Artemisiae Annuae Flos Lonicerae precipitate with ethanol process clearance standard the most in real time method for building up and clearance method and application |
| CN106323909A (en) * | 2016-09-14 | 2017-01-11 | 江苏大学 | Handheld near infrared spectrum detection system and detection method for quality of fruits and vegetables |
| CN106841103A (en) * | 2017-03-01 | 2017-06-13 | 沈阳农业大学 | Near infrared spectrum detects fruit internal quality method and dedicated test system |
| CN108519348A (en) * | 2018-04-17 | 2018-09-11 | 宁夏医科大学 | Licorice medicinal materials Near-Infrared Quantitative Analysis model and detection method and standard |
| CN112684023A (en) * | 2020-12-02 | 2021-04-20 | 太极集团重庆涪陵制药厂有限公司 | Rapid detection method for quality of magnolia officinalis medicinal material and screening method for magnolia officinalis medicinal material |
-
2022
- 2022-07-18 CN CN202210842698.6A patent/CN115128036A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104237060A (en) * | 2014-10-05 | 2014-12-24 | 浙江大学 | Multi-index quick detection method of honeysuckle |
| CN104833651A (en) * | 2015-04-15 | 2015-08-12 | 浙江大学 | Honeysuckle concentration process online real-time discharging detection method |
| CN104792652A (en) * | 2015-05-02 | 2015-07-22 | 浙江大学 | Multi-index rapid detection method for radix astragali |
| CN106226264A (en) * | 2016-05-05 | 2016-12-14 | 江苏康缘药业股份有限公司 | Herba Artemisiae Annuae Flos Lonicerae precipitate with ethanol process clearance standard the most in real time method for building up and clearance method and application |
| CN106323909A (en) * | 2016-09-14 | 2017-01-11 | 江苏大学 | Handheld near infrared spectrum detection system and detection method for quality of fruits and vegetables |
| CN106841103A (en) * | 2017-03-01 | 2017-06-13 | 沈阳农业大学 | Near infrared spectrum detects fruit internal quality method and dedicated test system |
| CN108519348A (en) * | 2018-04-17 | 2018-09-11 | 宁夏医科大学 | Licorice medicinal materials Near-Infrared Quantitative Analysis model and detection method and standard |
| CN112684023A (en) * | 2020-12-02 | 2021-04-20 | 太极集团重庆涪陵制药厂有限公司 | Rapid detection method for quality of magnolia officinalis medicinal material and screening method for magnolia officinalis medicinal material |
Non-Patent Citations (4)
| Title |
|---|
| 乔延江;杜敏;史新元;吴志生;徐冰;: "欧盟制药工业近红外光谱技术应用、申报和变更资料要求指南(草案)", 世界科学技术(中医药现代化), no. 04, 20 August 2012 (2012-08-20) * |
| 吴莎;刘启安;吴建雄;靳瑞婷;孙仙玲;刘茜;毕宇安;王振中;萧伟;: "统计过程控制结合近红外光谱在栀子中间体纯化工艺过程批放行中的应用研究", 中草药, no. 14, 28 July 2015 (2015-07-28) * |
| 孙飞;徐冰;戴胜云;史新元;乔延江;: "近红外分析用于中药产品质量实时放行测试的可靠性研究", 中华中医药杂志, no. 12, 1 December 2017 (2017-12-01) * |
| 谢兰桂: "再生塑料的识别研究", 《中国博士学位论文全文数据库 (工程科技Ⅱ辑)》, no. 01, 31 January 2013 (2013-01-31), pages 028 - 3 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116165165A (en) * | 2023-04-25 | 2023-05-26 | 四川威斯派克科技有限公司 | Detection method for online real-time release of raw and auxiliary materials of medicines |
| CN116660206A (en) * | 2023-05-31 | 2023-08-29 | 浙江省农业科学院 | Crop yield estimation method and system |
| CN116660206B (en) * | 2023-05-31 | 2024-05-28 | 浙江省农业科学院 | Crop yield estimation method and system |
| CN116660185A (en) * | 2023-06-02 | 2023-08-29 | 成都信息工程大学 | Multi-wavelength heavy metal ion solution detection system and detection method thereof |
| CN116660185B (en) * | 2023-06-02 | 2023-12-08 | 成都信息工程大学 | Multi-wavelength heavy metal ion solution detection system and detection method thereof |
| CN116953488A (en) * | 2023-09-19 | 2023-10-27 | 深圳市东陆科技有限公司 | Monitoring method for integrated photoelectric chip |
| CN116953488B (en) * | 2023-09-19 | 2023-12-12 | 深圳市东陆科技有限公司 | Monitoring method for integrated photoelectric chip |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115128036A (en) | Handheld near infrared spectrum detection method and device | |
| JP6375557B2 (en) | System and method for measuring and categorizing surface colors and spectra | |
| CN107860473A (en) | A kind of intelligent handhold spectrometer based on cloud data | |
| CN105466881A (en) | Portable near-infrared spectrum detection system | |
| CN105651728B (en) | A kind of portable near infrared spectrum detection system | |
| CN205352932U (en) | Portable pesticide residue detecting system | |
| CN103281459A (en) | Mobile phone capable of measuring sweetness and PH value of fruit | |
| CN105527246A (en) | Portable near infrared spectrum detection system | |
| CN106932360A (en) | Portable near infrared spectrum food science literature and modeling integral system and method | |
| CN108007912B (en) | Spectrum correction analysis device, correction analysis method, and drug authenticity determination system | |
| CN102692397A (en) | Portable optical fiber surface plasmons resonance (SPR) food safety detector | |
| CN113679355A (en) | Skin complexion measuring device and method for intense pulse light xerophthalmia treatment | |
| CN203337259U (en) | Hand-held optical radiation meter | |
| CN108931491A (en) | A kind of Multifunctional water portable remote monitor and water quality monitoring method | |
| CN209802980U (en) | near infrared spectrum analyzer | |
| CN207779882U (en) | A kind of peanut quality non-destructive testing device | |
| CN110118766A (en) | Hand-held general matter analysis system | |
| CN205562381U (en) | Portable near infrared spectroscopy detecting system | |
| CN108303379A (en) | The potato detection device of hand-held | |
| CN209961689U (en) | Portable test paper strip scanning and reading instrument | |
| CN208171857U (en) | Hand-held near-infrared spectrum detection device and near infrared spectrum detection system | |
| CN113406038A (en) | Optical detection method and device for pH value of water | |
| CN205538665U (en) | Portable near infrared spectroscopy detecting system | |
| CN209802978U (en) | remote real-time online fruit quality monitoring system | |
| CN209231204U (en) | A kind of laboratory test Fourier infrared spectrograph |
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 |