CN111401794A - Feed quality control method based on near infrared spectrum - Google Patents
Feed quality control method based on near infrared spectrum Download PDFInfo
- Publication number
- CN111401794A CN111401794A CN202010333657.5A CN202010333657A CN111401794A CN 111401794 A CN111401794 A CN 111401794A CN 202010333657 A CN202010333657 A CN 202010333657A CN 111401794 A CN111401794 A CN 111401794A
- Authority
- CN
- China
- Prior art keywords
- raw material
- prediction model
- samples
- spectrum
- product
- 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
- 238000003908 quality control method Methods 0.000 title claims abstract description 24
- 238000002329 infrared spectrum Methods 0.000 title claims abstract description 18
- 239000002994 raw material Substances 0.000 claims abstract description 115
- 238000001228 spectrum Methods 0.000 claims abstract description 65
- 238000000034 method Methods 0.000 claims abstract description 40
- 238000004519 manufacturing process Methods 0.000 claims abstract description 8
- 238000007689 inspection Methods 0.000 claims abstract description 6
- 238000005070 sampling Methods 0.000 claims abstract description 6
- 238000012544 monitoring process Methods 0.000 claims abstract description 5
- 238000012545 processing Methods 0.000 claims description 50
- 238000012795 verification Methods 0.000 claims description 23
- 238000001514 detection method Methods 0.000 claims description 16
- 238000012937 correction Methods 0.000 claims description 12
- 238000011156 evaluation Methods 0.000 claims description 11
- 230000003595 spectral effect Effects 0.000 claims description 11
- 239000000126 substance Substances 0.000 claims description 9
- 238000010606 normalization Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 8
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 238000007621 cluster analysis Methods 0.000 claims description 6
- 230000005856 abnormality Effects 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 4
- 239000008188 pellet Substances 0.000 claims description 3
- 238000002790 cross-validation Methods 0.000 claims 2
- 235000012054 meals Nutrition 0.000 claims 1
- 239000000523 sample Substances 0.000 description 32
- 239000000463 material Substances 0.000 description 7
- 239000000843 powder Substances 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000002452 interceptive effect Effects 0.000 description 4
- 239000008187 granular material Substances 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 235000019733 Fish meal Nutrition 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000004467 fishmeal Substances 0.000 description 1
- 238000005469 granulation Methods 0.000 description 1
- 230000003179 granulation Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 235000015097 nutrients Nutrition 0.000 description 1
- 235000016709 nutrition Nutrition 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
-
- 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
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- 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
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Analytical Chemistry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Pathology (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
- Manufacturing & Machinery (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention relates to a feed quality control method based on near infrared spectrum, which comprises the following steps: modeling: establishing a corresponding prediction model by using the raw materials and products with the concentration range set by an enterprise, wherein the prediction model comprises a raw material prediction model and a product prediction model; raw material quality control: obtaining a characteristic spectrum of the raw material by using a raw material prediction model, collecting related information in the characteristic spectrum, and performing raw material production area identification, similar unknown sample inspection, raw material abnormity judgment and/or raw material stability judgment; and (3) controlling the product quality: and (4) sampling, scanning and comparing each mixing unit and the output product in the feed by using a product prediction model, and monitoring the product quality in real time. Compared with the prior art, the method has the advantages of reducing the modeling time of the prediction model, dynamically adjusting the width and the adaptability of the model, improving the effectiveness of feed quality control, along with high matching degree, wide applicability, improvement on the use value of the near-infrared detector and the like.
Description
Technical Field
The invention relates to the field of feed quality detection, in particular to a feed quality control method based on near infrared spectrum.
Background
The near infrared spectrum technology for measuring the chemical composition and the physicochemical property of the sample has the advantages of high efficiency, environmental protection and the like, and feed enterprises in recent years also widely use a near infrared rapid detection mode to perfect the quality control. In order to facilitate users to use the near-infrared analyzers more efficiently, near-infrared analyzer manufacturers provide near-infrared networked management services, namely, a plurality of near-infrared analyzers are connected into an integral data analysis system by using the internet, a server or other modes, but the problems of incomplete prediction models, insufficient data spectrum utilization and the like still exist.
Chinese patent CN201510490572.7 discloses a method for controlling feed production on line by using a near infrared technology, wherein the feed production is controlled on line by using the near infrared technology, a near infrared spectrometer detects near infrared spectrum information of feed raw material components and content information of nutrient components in a feed hopper through a near infrared probe and transmits the information to a computer provided with a formula system, and computer software judges the quality of the feed raw materials according to the obtained information and grades the feed raw materials according to the judgment result. The method can realize the on-site nondestructive rapid analysis of raw materials, process products and final products involved in the feed production process by utilizing a near infrared spectrum analysis technology, but a prediction model is established according to near infrared spectrum information of a formula which can be used as feed raw materials in a formula system, the modeling method needs a large amount of sample data, the collection of samples is difficult for a single enterprise, the working pressure of testing and testing is extremely high, the method only analyzes the quality of the feed raw materials and the nutritional value of the raw materials, other quality control methods utilizing the near infrared spectrum are not provided, and the method cannot be suitable for the quality control of various different feed enterprises.
At present, a single index or a single product prediction model is basically established, a great number of samples need to be collected in the process of establishing the prediction model, although the model is relatively wide in applicability, the workload of testing and testing is very large, and in combination with the current situation, the common phenomenon is that the raw materials and the varieties and formulas of a common enterprise are various, so that the laboratory detection capability needed to be equipped is also quite high. The modeling method provided at present has huge workload for enterprises, generally, the expensive near-infrared detectors are difficult to be put into quality control in a short time, and a method for maximizing the value of the enterprises to the near-infrared detectors is lacked. Therefore, how to improve the working efficiency and the value of the near-infrared feed detector is a technical problem to be solved by feed enterprises nowadays.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a feed quality control method based on near infrared spectrum, which reduces modeling time.
The purpose of the invention can be realized by the following technical scheme:
a feed quality control method based on near infrared spectrum comprises the following steps:
modeling: establishing a corresponding prediction model by using the raw materials and products with the concentration range set by an enterprise, wherein the prediction model comprises a raw material prediction model and a product prediction model;
raw material quality control: obtaining a characteristic spectrum of the raw material by using a raw material prediction model, collecting related information in the characteristic spectrum, and performing raw material production area identification, similar unknown sample inspection, raw material abnormity judgment and/or raw material stability judgment;
and (3) controlling the product quality: and (4) sampling, scanning and comparing each mixing unit and the output product in the feed by using a product prediction model, and monitoring the product quality in real time.
The calibration software adopts WinISI calibration software.
Further, the modeling step specifically includes:
s101: performing model verification on an existing raw material prediction model provided by an instrument provider, and performing model expansion according to a verification result;
s102: collecting raw material samples and product samples required by modeling, and numbering and crushing the raw material samples and the product samples according to set requirements;
s103: scanning the processed raw material sample and the product sample to obtain a spectrum and inputting the spectrum into calibration software;
s104: the calibration software screens out samples which accord with a set concentration range according to the spectral data, and wet chemical detection is carried out on the samples;
s105: the calibration software combines the spectral data with the wet chemical detection result, selects an optimal processing mode through different mathematical processing evaluations, and establishes a prediction model;
s106: and regularly carrying out model verification on the established prediction model, and carrying out model expansion according to a verification result.
Further preferably, the model verification specifically includes: selecting 5-10 samples with different concentrations, verifying the prediction accuracy of the corresponding prediction model, and gradually and dynamically adjusting the width and the adaptability of the model in the later stage;
the model extension specifically includes: and expanding the index concentration range of the prediction model with an unsatisfactory verification result, and performing raw material collection, inspection and test, scanning and calibration again.
The prediction model with an unsatisfactory verification result refers to a prediction model with a scanning alarm or a prediction result and a scanning value exceeding an allowable error, wherein the allowable error is an industry standard.
Further preferably, in step S104, the number of samples that are screened to meet the set concentration range is in the range of 30-50.
Further preferably, the mathematical processing evaluation comprises a calibration standard deviation and a calibration correlation coefficient, an interactive validation standard deviation and an interactive validation correlation coefficient;
the processing mode comprises a scattering correction method and derivative mathematical processing, the scattering correction method comprises normalization, deviation removal, normalization and deviation removal, multivariate scattering correction and no processing, and the derivative mathematical processing comprises a first derivative and a second derivative;
the method for selecting the optimal processing mode through different mathematical processing evaluations specifically comprises the following steps: and selecting a mode with the minimum standard deviation and the maximum correlation coefficient as an optimal processing mode by selecting different scattering correction methods and different derivative mathematical processing.
Further, the identification of the origin of the raw materials specifically comprises:
s201: collecting raw material samples with reliable sample source producing areas, and determining the characteristics of the sample producing areas by combining microscopic examination;
s202: scanning a raw material sample to be detected, and collecting a characteristic spectrum of the raw material sample through a prediction model;
s203: the calibration software calculates scores by using the spectral data only through cluster analysis and generates a three-dimensional map of the producing area;
s204: counting the difference among the samples according to the three-dimensional map of the producing area, acquiring a characteristic spectrum set, and identifying the producing area of unknown raw materials;
the similar unknown sample detection specifically comprises the following steps:
s211: collecting the same raw material sample with known source and index gradient, and collecting the characteristic spectrum;
s212: the calibration software calculates scores by using the spectral data only and generates a classification three-dimensional graph through clustering analysis;
s213: and counting the difference among the index gradient samples according to the classified three-dimensional graph, acquiring a characteristic spectrum set, and detecting the similar unknown samples.
Further preferably, in step S203 and step S212, the cluster analysis uses a PCA method.
Further, the raw material abnormality determination specifically includes:
s221: collecting a raw material sample with a known source, and collecting a characteristic spectrum of the raw material sample;
s222: the calibration software performs mathematical processing by using the characteristic spectrum acquired in step S221 to obtain a characteristic spectrum set thereof;
s223: collecting a characteristic spectrum of an unknown sample, and obtaining a comparison spectrum by adopting a spectrum processing mode in the step S222;
s224: comparing the comparison spectrum with the characteristic spectrum set to identify whether the unknown raw material is abnormal or not;
the raw material stability judgment specifically comprises the following steps:
s231: scanning raw material samples of each batch and collecting characteristic spectra of the raw material samples;
s232: the calibration software utilizes the characteristic spectrum collected in the step S231 to obtain a characteristic spectrum set after mathematical processing;
s233: comparing the characteristic spectrum set of each batch of raw material samples to judge the stability of the raw materials.
Further preferably, in step S222 and step S232, the calibration software obtains the characteristic spectrum set after using the de-scattering and normalization and the mathematical processing of the first derivative.
Furthermore, the product prediction model comprises a particle prediction model and a powder prediction model. For granulated feed, various raw material powder materials are mixed and then are subjected to modulation and granulation to obtain granulated materials, the granulated feed has two forms of powder materials and granules, the powder materials are taken out in a mixing unit, and the granulated materials are taken out at a discharge port, so that when the product is the granulated feed, a granule prediction model and a powder material prediction model need to be established at the same time to improve the accuracy of prediction data.
Compared with the prior art, the invention has the following advantages:
1) the method utilizes the concentration ranges of the raw materials and the products of the enterprises to carry out modeling, generates the prediction models suitable for the respective enterprises, reduces the workload of the samples and laboratories involved in modeling, ensures that the enterprises can build the suitable prediction models in the shortest time and put into quality control work, and has high matching degree and wide applicability;
2) according to the method, by processing and comparing different methods and objects of the spectrum, operations such as identification of unknown raw material producing areas, detection of similar unknown samples, abnormal judgment of raw materials, stability judgment of raw materials and the like can be performed, quality control is performed on the raw materials from multiple aspects, and the effectiveness of feed quality control is improved;
3) according to the method, the sampling and stability monitoring of the granulated feed product are realized by establishing the granulated prediction model and the powder prediction model, and a part of online near-infrared functions are achieved, so that an enterprise can well monitor the product quality under the condition of not putting into online near-infrared, and the value maximization of an enterprise near-infrared detector is achieved.
Drawings
FIG. 1 is a schematic diagram of a predictive model building process;
FIG. 2 is a diagram showing an example of the identification result of the origin of the raw material;
FIG. 3 is a diagram showing an example of the test results of the same type of unknown sample;
FIG. 4 is a diagram showing an example of the determination result of abnormal material;
fig. 5 is a diagram showing an example of the determination result of the stability of the raw material.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, shall fall within the scope of protection of the present invention.
Examples
The invention provides a feed enterprise quality control method based on near infrared spectrum, which comprises the following steps:
modeling: establishing a corresponding prediction model by utilizing the raw materials and products with the concentration range set by an enterprise, wherein the prediction model comprises a raw material prediction model and a product prediction model;
raw material quality control: obtaining a characteristic spectrum of the raw material by using the raw material prediction model and inputting the characteristic spectrum into calibration software, wherein the calibration software acquires relevant information in the characteristic spectrum and performs raw material production area identification, similar unknown sample inspection, raw material abnormity judgment and/or raw material stability judgment;
and (3) controlling the product quality: and (4) sampling, scanning and comparing each mixing unit and the output product in the feed by using a product prediction model, and monitoring the product quality in real time.
As shown in fig. 1, the modeling step mainly includes the following steps:
s101: performing model verification on an existing raw material prediction model provided by an instrument provider, and performing model expansion according to a verification result;
s102: collecting raw material samples and product samples required by modeling, and numbering and crushing the raw material samples and the product samples according to set requirements;
s103: scanning the processed raw material sample and the product sample to obtain a spectrum and inputting the spectrum into calibration software;
s104: the calibration software screens out samples which accord with a set concentration range according to the spectral data, and wet chemical detection is carried out on the samples;
s105: the calibration software combines the spectral data with the wet chemical detection result, selects an optimal processing mode through different mathematical processing evaluations, and establishes a prediction model;
the mathematical processing evaluation comprises a calibration standard deviation, a calibration correlation coefficient, an interactive verification standard deviation and an interactive verification correlation coefficient;
the processing mode comprises a scattering correction method and derivative mathematical processing, wherein the scattering correction method comprises normalization, deviation removal, normalization and deviation removal, multivariate scattering correction and no processing, and the derivative mathematical processing comprises a first derivative and a second derivative;
the method for selecting the optimal processing mode through different mathematical processing evaluations specifically comprises the following steps: and selecting a mode with the minimum standard deviation and the maximum correlation coefficient as an optimal processing mode by selecting different scattering correction methods and different derivative mathematical processing.
S106: and regularly carrying out model verification on the established prediction model, and carrying out model expansion according to a verification result.
In this embodiment, the overall process of establishing the prediction model is as follows:
1) according to the instruction of the detector and the results of the enterprise products, normative documents such as an instrument operation manual, a specified sample number rule, a sample crushing granularity and the like are formulated;
2) arranging the raw materials, product catalogues, raw material acceptance standards and product delivery indexes of an enterprise, determining and establishing a necessary raw material prediction model, a product prediction model and approximate workload, and making a model establishment plan;
3) for common raw material prediction models provided by part of instrument manufacturers, making model verification and extension plans, and generally selecting 5-10 samples with different concentrations to verify the model prediction accuracy;
4) collecting required raw material samples and product samples according to a formulated plan, processing the samples according to the requirements of specified sample numbering rules and sample crushing granularity, and scanning the samples to obtain spectra;
5) processing the spectrum by using software to screen out samples with a certain concentration range, firstly establishing 30-50 samples covering different concentration indexes for small-range use of an enterprise, gradually and dynamically adjusting the width and the adaptability of the model in the later period, and carrying out wet chemical detection on the part of samples;
6) combining spectral data with wet chemical detection results by using WinISI calibration software, selecting an optimal processing mode through different mathematical processing evaluations, establishing a prediction model, mainly adopting de-scattering and standardization processing and adopting first-order derivative processing to establish the prediction model for a conventional feed project;
7) and (3) regularly verifying the model, making a model expansion plan according to the result of the verification model, continuously accumulating sample data in the actual detection process to perfect a prediction model, and performing model expansion on the model with an unsatisfactory verification result, wherein the prediction model with the unsatisfactory verification result refers to a prediction model with a scanning alarm or a prediction result and a scanning value exceeding an allowable error, and the allowable error is an industrial standard.
In this embodiment, in the step of controlling the quality of the raw material, the processes of identifying the origin of the raw material, inspecting an unknown sample of the same type, determining the abnormality of the raw material, and determining the stability of the raw material are specifically as follows:
identifying the origin of the raw materials: collecting reliable samples of a sample source place, and determining the characteristics of the sample source place by combining microscopic examination; scanning a sample to be detected and collecting a characteristic spectrum of the sample; WinISI calibration software adopts a PCA mode in a cluster analysis method, only scanning data in a file is used for calculating scores and making a three-dimensional map of a producing area, so that the difference among all samples is counted, a characteristic spectrum set is obtained, the producing area of unknown raw materials is identified, and the result is shown in figure 2, which is the identification result of the producing area of the fish meal raw materials.
And (3) testing similar unknown samples: collecting the same type of raw material samples with known sources and index gradients, and collecting characteristic spectra of the raw material samples; WinISI calibration software adopts a PCA mode in a cluster analysis method, only uses scanning data in a file to calculate scores and make a classified three-dimensional graph, and thus, the difference between gradient samples of each index is counted, so that a characteristic spectrum set is obtained and is used for rapidly detecting similar unknown samples, and the result is shown in figure 3.
And (3) judging the abnormality of the raw materials: the method comprises the steps of collecting a spectrum obtained by scanning a certain raw material sample with a known source, collecting a characteristic spectrum set of the raw material sample by WinISIS calibration software in a de-scattering and standardization and first-order derivative mathematical processing mode, scanning an unknown sample to obtain a spectrum, obtaining a comparison spectrum in the same spectrum processing mode, and rapidly identifying whether the unknown raw material is abnormal or not, wherein the result is shown in figure 4, and a quality report can be obtained by combining the abnormal spectrum with microscopic examination.
And (3) judging the stability of the raw materials: scanning raw material samples of the same manufacturer in each batch, collecting characteristic spectrums of the raw material samples, collecting characteristic spectrum sets of the raw material samples by adopting a de-scattering and standardization mode and a first-order derivative mathematical processing mode through WinISIS calibration software, and quickly judging the stability of the raw material, wherein the result is shown in figure 5 and can provide important basis for evaluation of suppliers.
In this embodiment, in the product quality control step, a pellet prediction model and a powder prediction model are respectively established for the same detection data as for pellet feeds, sampling, scanning and comparing are performed for each mixing unit and the produced product, the product quality is monitored in real time, the stability of the quality of the produced product is ensured, and meanwhile, various product quality reports are rapidly generated by using the scanning data, so that formula adjustment and production guidance are facilitated.
While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and those skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A feed quality control method based on near infrared spectrum is characterized by comprising the following steps:
modeling: establishing a corresponding prediction model by using the raw materials and products with the concentration range set by an enterprise, wherein the prediction model comprises a raw material prediction model and a product prediction model;
raw material quality control: obtaining a characteristic spectrum of the raw material by using a raw material prediction model, collecting related information in the characteristic spectrum, and performing raw material production area identification, similar unknown sample inspection, raw material abnormity judgment and/or raw material stability judgment;
and (3) controlling the product quality: and (4) sampling, scanning and comparing each mixing unit and the output product in the feed by using a product prediction model, and monitoring the product quality in real time.
2. The feed quality control method based on near infrared spectrum according to claim 1, wherein the modeling step specifically comprises:
s101: performing model verification on an existing raw material prediction model provided by an instrument provider, and performing model expansion according to a verification result;
s102: collecting raw material samples and product samples required by modeling, and numbering and crushing the raw material samples and the product samples according to set requirements;
s103: scanning the processed raw material sample and the product sample to obtain a spectrum and inputting the spectrum into calibration software;
s104: the calibration software screens out samples which accord with a set concentration range according to the spectral data, and wet chemical detection is carried out on the samples;
s105: the calibration software combines the spectral data with the wet chemical detection result, selects an optimal processing mode through different mathematical processing evaluations, and establishes a prediction model;
s106: and regularly carrying out model verification on the established prediction model, and carrying out model expansion according to a verification result.
3. The method for controlling the quality of the feed based on the near infrared spectrum as claimed in claim 2, wherein the model verification specifically comprises: selecting 5-10 samples with different concentrations, and verifying the prediction accuracy of the corresponding prediction model;
the model extension specifically includes: and expanding the index concentration range of the prediction model with an unsatisfactory verification result, and performing raw material collection, inspection and test, scanning and calibration again.
4. The method as claimed in claim 2, wherein the number of samples selected in step S104 to meet the predetermined concentration range is 30-50.
5. The near infrared spectrum-based feed quality control method according to claim 2, wherein the mathematical processing evaluation comprises a calibration standard deviation and a calibration correlation coefficient, a cross validation standard deviation and a cross validation correlation coefficient;
the processing mode comprises a scattering correction method and derivative mathematical processing, the scattering correction method comprises normalization, deviation removal, normalization and deviation removal, multivariate scattering correction and no processing, and the derivative mathematical processing comprises a first derivative and a second derivative;
the method for selecting the optimal processing mode through different mathematical processing evaluations specifically comprises the following steps: and selecting a mode with the minimum standard deviation and the maximum correlation coefficient as an optimal processing mode by selecting different scattering correction methods and different derivative mathematical processing.
6. The method as claimed in claim 1, wherein the identification of the source location of the raw material comprises:
s201: collecting raw material samples with reliable sample source producing areas, and determining the characteristics of the sample producing areas by combining microscopic examination;
s202: scanning a raw material sample to be detected, and collecting a characteristic spectrum of the raw material sample through a prediction model;
s203: the calibration software calculates scores by using the spectral data only through cluster analysis and generates a three-dimensional map of the producing area;
s204: counting the difference among the samples according to the three-dimensional map of the producing area, acquiring a characteristic spectrum set, and identifying the producing area of unknown raw materials;
the similar unknown sample detection specifically comprises the following steps:
s211: collecting the same raw material sample with known source and index gradient, and collecting the characteristic spectrum;
s212: the calibration software calculates scores by using the spectral data only and generates a classification three-dimensional graph through clustering analysis;
s213: and counting the difference among the index gradient samples according to the classified three-dimensional graph, acquiring a characteristic spectrum set, and detecting the similar unknown samples.
7. The method as claimed in claim 6, wherein the PCA is used in the cluster analysis in step S203 and step S212.
8. The method for controlling the quality of the feed based on the near infrared spectrum as claimed in claim 1, wherein the determining of the abnormality of the raw material specifically comprises:
s221: collecting a raw material sample with a known source, and collecting a characteristic spectrum of the raw material sample;
s222: the calibration software performs mathematical processing by using the characteristic spectrum acquired in step S221 to obtain a characteristic spectrum set thereof;
s223: collecting a characteristic spectrum of an unknown sample, and obtaining a comparison spectrum by adopting a spectrum processing mode in the step S222;
s224: comparing the comparison spectrum with the characteristic spectrum set to identify whether the unknown raw material is abnormal or not;
the raw material stability judgment specifically comprises the following steps:
s231: scanning raw material samples of each batch and collecting characteristic spectra of the raw material samples;
s232: the calibration software utilizes the characteristic spectrum collected in the step S231 to obtain a characteristic spectrum set after mathematical processing;
s233: comparing the characteristic spectrum set of each batch of raw material samples to judge the stability of the raw materials.
9. The method as claimed in claim 8, wherein the calibration software in steps S222 and S232 obtains the set of characteristic spectra after using the de-scattering and normalization and the mathematical first derivative.
10. The near infrared spectrum-based feed quality control method of claim 1, wherein the product prediction model comprises a pellet prediction model and a meal prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010333657.5A CN111401794A (en) | 2020-04-24 | 2020-04-24 | Feed quality control method based on near infrared spectrum |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010333657.5A CN111401794A (en) | 2020-04-24 | 2020-04-24 | Feed quality control method based on near infrared spectrum |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111401794A true CN111401794A (en) | 2020-07-10 |
Family
ID=71429802
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010333657.5A Pending CN111401794A (en) | 2020-04-24 | 2020-04-24 | Feed quality control method based on near infrared spectrum |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111401794A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113588596A (en) * | 2021-08-25 | 2021-11-02 | 浙江大学 | Method and system for identifying and detecting content of animal source protein powder doped in fish meal |
| CN115452763A (en) * | 2022-10-27 | 2022-12-09 | 佛山市南海区杰大饲料有限公司 | Functional fish feed detection method and system based on near infrared spectrum |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102564993A (en) * | 2011-12-31 | 2012-07-11 | 江南大学 | Method for identifying rice varieties by using Fourier transform infrared spectrum and application of method |
| CN104777130A (en) * | 2015-04-21 | 2015-07-15 | 通威股份有限公司 | Near infrared spectrum technique-based rapid detection method for L-threonine |
| CN105424675A (en) * | 2015-11-05 | 2016-03-23 | 中国农业大学 | Ruminant animal-origin feedstuff identification method based on lipid Raman spectrums |
| CN110209141A (en) * | 2019-07-12 | 2019-09-06 | 青海盐湖工业股份有限公司 | A kind of quality testing management system |
-
2020
- 2020-04-24 CN CN202010333657.5A patent/CN111401794A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102564993A (en) * | 2011-12-31 | 2012-07-11 | 江南大学 | Method for identifying rice varieties by using Fourier transform infrared spectrum and application of method |
| CN104777130A (en) * | 2015-04-21 | 2015-07-15 | 通威股份有限公司 | Near infrared spectrum technique-based rapid detection method for L-threonine |
| CN105424675A (en) * | 2015-11-05 | 2016-03-23 | 中国农业大学 | Ruminant animal-origin feedstuff identification method based on lipid Raman spectrums |
| CN110209141A (en) * | 2019-07-12 | 2019-09-06 | 青海盐湖工业股份有限公司 | A kind of quality testing management system |
Non-Patent Citations (1)
| Title |
|---|
| 陈朋: "近红外分析技术在银蓝调脂胶囊制备过程中的应用研究" * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113588596A (en) * | 2021-08-25 | 2021-11-02 | 浙江大学 | Method and system for identifying and detecting content of animal source protein powder doped in fish meal |
| CN115452763A (en) * | 2022-10-27 | 2022-12-09 | 佛山市南海区杰大饲料有限公司 | Functional fish feed detection method and system based on near infrared spectrum |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102590129B (en) | Method for detecting content of amino acid in peanuts by near infrared method | |
| CN105717066B (en) | A kind of near infrared spectrum identification model based on weighted correlation coefficient | |
| Febbi et al. | Automated determination of poplar chip size distribution based on combined image and multivariate analyses | |
| CN111401794A (en) | Feed quality control method based on near infrared spectrum | |
| CN108760647A (en) | A kind of wheat content of molds line detecting method based on Vis/NIR technology | |
| CN104297206A (en) | Quick liquid milk brand identifying method | |
| CN105223140A (en) | The method for quickly identifying of homology material | |
| CN100557413C (en) | Ammonium perchlorate particle size's rapid analysis in the solid propellant | |
| CN110210005A (en) | A kind of spectrum wave number selection method of no reference value | |
| CN108663334B (en) | Method for searching spectral characteristic wavelength of soil nutrient based on multi-classifier fusion | |
| CN109283153B (en) | Method for establishing quantitative analysis model of soy sauce | |
| CN108169168A (en) | Test and analyze rice grain protein content mathematical model and construction method and application | |
| CN116952895A (en) | Method for rapidly determining mixing uniformity | |
| CN113655027A (en) | Method for rapidly detecting tannin content in plant by near infrared | |
| CN105675538A (en) | Method for detection of linseed cake nutrients | |
| Lodder et al. | Quantile analysis: a method for characterizing data distributions | |
| CN106053380A (en) | Method for quickly analyzing proportions of tree species of mixed pulp wood by aid of near infrared spectrum technologies | |
| CN107238557A (en) | A kind of method of utilization near infrared spectroscopy quick detection calcium carbonate particle diameter distribution | |
| WO2003056300A2 (en) | Systems and methods for automated quantitative analyses of digitized spectra | |
| CN104181125A (en) | Method for rapidly determining Kol-bach value of beer malt | |
| Foroozani et al. | Classification of wheat varieties by PLS-DA and LDA models and investigation of the spatial distribution of protein content using NIR spectroscopy. | |
| CN111257265A (en) | Rapid detection method for penetration index of asphalt for steel bridge deck pavement | |
| CN113834795A (en) | Hydroxychloroquine sulfate particle moisture near infrared spectrum online quantitative model and establishing method and detection method thereof | |
| CN1712937A (en) | Nondestructive rapid detection for contents of multiple element components in composite fertilizer | |
| JP2009210445A (en) | Production and inspection monitor system, autoanalyzer and data processing method of them |
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 |