CN119619020A

CN119619020A - Method and system for detecting impurities in the production process of glucosamine

Info

Publication number: CN119619020A
Application number: CN202411714755.8A
Authority: CN
Inventors: 王行义; 徐新
Original assignee: Jiyuan Biological Products Xuzhou Co ltd
Current assignee: Jiyuan Biological Products Xuzhou Co ltd
Priority date: 2024-11-27
Filing date: 2024-11-27
Publication date: 2025-03-14
Anticipated expiration: 2044-11-27
Also published as: CN119619020B

Abstract

The present invention discloses an impurity detection method and system in the glucosamine production process, and relates to the field of detection technology. The method comprises: obtaining quality control samples of multiple steps by sampling, extracting impurity type sets in combination with production logs, and configuring impurity detection models and call tables. The sample impurities are analyzed for cumulative regularity, key steps and their corresponding impurity types are located, and based on this, process parameters are optimized and an online detection module is configured. The optimized process parameters are applied in the key steps, and dynamic monitoring and early warning of impurity deviations are achieved through the online detection module. The technical problem of insufficient impurity control in the existing glucosamine production process is solved, and the accuracy and efficiency of impurity control in glucosamine production are improved, and the technical effect of ensuring the consistency and stability of product quality is achieved.

Description

Impurity detection method and system in glucosamine production process

Technical Field

The application relates to the technical field of detection, in particular to a method and a system for detecting impurities in a glucosamine production process.

Background

Glucosamine is an important raw material for medical and health care products, and the stability of the purity and quality of the product in the production process is critical to the safety and effectiveness of the final application. However, the accumulation, conversion or retention of various impurities during the production process often results in difficulty in achieving high standard purity of the final product. Most of the prior art adopts single-point detection and off-line analysis means, can not fully reflect the change rule of impurities in the production process, is difficult to respond in time, and affects production optimization and control. Especially in multi-step production, it is still a difficult problem to precisely locate the critical process steps and their corresponding impurity species to achieve efficient impurity removal. Therefore, a real-time monitoring and optimizing control method combined with production process data is needed to improve the product quality and the production efficiency and reduce the production cost.

In the related art at the present stage, the technical problem of insufficient impurity control in the glucosamine production process exists.

Disclosure of Invention

The application solves the technical problem of insufficient impurity control in the existing glucosamine production process by providing the method and the system for detecting the impurities in the glucosamine production process.

The application provides an impurity detection method in the production process of glucosamine, which comprises the following steps:

Sampling in a glucosamine test production process to obtain a plurality of quality control samples of a plurality of production steps, wherein the plurality of production steps are provided with a plurality of execution sequence identifications, after a sample impurity type set is obtained through an interactive production log, carrying out impurity detection model configuration according to the sample impurity type set to obtain a detection model call table, carrying out impurity accumulation rule analysis on the plurality of quality control samples, positioning K groups of key impurity types of K key steps based on analysis results, wherein the K key steps are provided with K execution sequence identifications, carrying out coupling optimization on the K key steps according to the K groups of key impurity types as impurity removal guide, outputting K optimized process parameters, carrying out on-line detection function configuration on the K key steps according to the K groups of key impurity types and the detection model call table to obtain K on-line detection modules, and carrying out deviation dynamic detection early warning on impurities in the glucosamine production process through the K on-line detection modules after carrying out process parameter updating on the K key steps by adopting the K optimized process parameters.

The application also provides an impurity detection system in the glucosamine production process, which comprises the following steps:

The system comprises a quality control sample acquisition module, an accumulation rule analysis module, a coupling optimization module, an online detection function configuration module, an early warning module and a dynamic early warning module, wherein the quality control sample acquisition module is used for acquiring a plurality of quality control samples of a plurality of production steps by sampling in a glucosamine test production process, the plurality of production steps are provided with a plurality of execution sequence identifiers, the impurity detection module is used for carrying out impurity detection module configuration according to a sample impurity type set after an interactive production log acquires the sample impurity type set, the accumulation rule analysis module is used for carrying out impurity accumulation rule analysis on the plurality of quality control samples and positioning K groups of key impurity types of K key steps based on analysis results, the K key steps are provided with K execution sequence identifiers, the coupling optimization module is used for carrying out coupling optimization on the K key steps according to the K execution sequence identifiers and outputting K optimized process parameters, the online detection function configuration module is used for carrying out on-line detection on the K key steps according to the K groups of impurity type and the detection rule analysis module, and the dynamic early warning module is used for carrying out dynamic early warning on-line detection on the K key steps by adopting the dynamic glucose test process after the K key steps are dynamically optimized.

According to the method and the system for detecting the impurities in the glucosamine production process, quality control samples of a plurality of steps are obtained through sampling, an impurity type set is extracted by combining production logs, and an impurity detection model and a call table are configured. And analyzing the accumulation rule of the sample impurities, positioning key steps and the impurity types corresponding to the key steps, optimizing process parameters based on the key steps, and configuring an online detection module. The optimized process parameters are applied in the key steps, and the dynamic monitoring and early warning of impurity deviation are realized through the online detection module, so that the technical effects of improving the accuracy and efficiency of impurity control in the glucosamine production and ensuring the consistency and stability of the product quality are achieved.

Drawings

In order to more clearly illustrate the technical solution of the embodiments of the present application, the following will briefly describe the drawings of the embodiments of the present application, in which flowcharts are used to illustrate operations performed by a system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in order precisely. Rather, the various steps may be processed in reverse order or simultaneously, as desired. Also, other operations may be added to or removed from these processes.

Fig. 1 is a schematic flow chart of an impurity detection method in a glucosamine production process according to an embodiment of the present application.

Fig. 2 is a schematic structural diagram of an impurity detection system in a glucosamine production process according to an embodiment of the present application.

The reference numerals illustrate the quality control sample acquisition module 10, the impurity detection model configuration module 20, the accumulation law analysis module 30, the coupling optimization module 40, the online detection function configuration module 50 and the dynamic detection early warning module 60.

Detailed Description

The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.

The present application will be described in further detail below with reference to the accompanying drawings, in order to make the objects, technical solutions and advantages of the present application more apparent, and the described embodiments should not be construed as limiting the present application, but all other embodiments obtained by those skilled in the art without making any inventive effort are within the scope of the present application.

In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is to be understood that "some embodiments" may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict, the term "first\second" being referred to merely as distinguishing between similar objects and not representing a particular ordering for the objects. The terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or server that comprises a list of steps or elements is not necessarily limited to those steps or elements that are expressly listed or inherent to such process, method, article, or apparatus, but may include other steps or modules that may not be expressly listed or inherent to such process, method, article, or apparatus, and unless otherwise defined, 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 application pertains. The terminology used herein is for the purpose of describing embodiments of the application only.

The embodiment of the application provides a method and a system for detecting impurities in a glucosamine production process, as shown in fig. 1, wherein the method comprises the following steps:

Step S100, obtaining a plurality of quality control samples of a plurality of production steps by sampling in a glucosamine test production process, wherein the plurality of production steps have a plurality of execution sequence identifications. Specifically, before glucosamine test production, a carding production flow is planned, each link is defined, and an execution sequence identifier is given. And determining sampling points according to the production steps and key links, and collecting liquid samples as quality control samples by using professional equipment in different stages such as after raw material treatment, during reaction, before and after intermediate product purification and after final product synthesis. The collected sample is marked with recorded information, including production step identification, sampling time, place, operators and the like, and file record appearance and primary physicochemical properties are established. And then, the sample is properly stored, and the pretreated reagent, instrument and equipment and operation rules are prepared according to the detection requirement, so that the sample is prevented from going bad or being disturbed, and the accuracy and comparability of the detection result are ensured.

And step S200, after the interactive production log obtains a sample impurity type set, carrying out impurity detection model configuration according to the sample impurity type set to obtain a detection model call table. The method comprises the steps of firstly, interacting glucosamine production logs, mining and sorting out sample impurity type sets, then integrating and classifying M sample impurity spectrum atlas containing M sample impurity content identification sets according to M sample impurity type networking call data, obtaining composite and single-dimensional impurity spectrograms by taking M sample impurity types as constraint networking, constructing M impurity content identification branches according to the spectrum atlas and the content identification sets by utilizing the construction layer separation models, connecting the layer separation models with the identification branches to form an impurity detection model, and finally associatively storing M sample impurity types and M impurity content identification branches to construct a detection model call table and storing so as to efficiently and accurately detect impurities in the production process.

In a possible implementation manner, after the sample impurity type set is obtained in the interactive production log, performing impurity detection model configuration according to the sample impurity type set to obtain a detection model call table, and step S200 further includes step S210 of performing networking data call according to M sample impurity types in the sample impurity type set to obtain M sample impurity spectrum atlas of the M sample impurity types, where the M sample impurity spectrum atlas has M sample impurity content identification sets. Specifically, a networking data calling program is started according to M sample impurity types in the sample impurity type set. The system is connected to a professional chemical database, a spectrum data sharing platform of a scientific research institution, a large data resource base in the industry and the like. In the massive data resources, the accurate search algorithm and the data screening technology are utilized to search and match one by one according to M sample impurity types. For example, if one of the sample impurity types is a metal ion impurity, the database is searched for spectral data of the metal ion under different chemical environments and different concentrations. And sorting and classifying the searched spectrum data corresponding to each sample impurity type, thereby obtaining M sample impurity spectrum atlas. And, in constructing these atlases, M sample impurity content identification sets are synchronously established. The quantitative relation between different spectral characteristics and specific impurity content is determined by analyzing and extracting a large amount of spectral data with accurate impurity content marks in a database, so that the impurity content can be reversely deduced according to the spectral characteristics when an unknown sample is detected later.

And S220, carrying out networking data call by taking the M sample impurity types as constraints to obtain a plurality of sample compound impurity spectrograms and a plurality of groups of sample single-dimensional impurity spectrograms. Specifically, M sample impurity types are taken as constraints to deeply enter a networking data resource, a plurality of spectral patterns are obtained in an emphasized mode, the spectral patterns comprise sample composite impurity spectral patterns reflecting the mixing condition of a plurality of impurities, spectral data containing the mixing condition of the known impurities are screened, the influence of impurity coexistence and interaction on spectral characteristics in the production of glucosamine is simulated, the spectral patterns of a plurality of groups of sample single-dimensional impurity are focused on the specific dimensional characteristics of single impurities, the data are derived from information under a specific wavelength range, polarization direction or spectral analysis technology, and the spectral characteristics of single impurities can be accurately presented by decomposing the composite spectral patterns or extracting the single impurity spectral resources due to the limitation of the impurities of the composite impurity spectrum.

And step S230, constructing and obtaining a layer separation model based on the multiple sample composite impurity spectrograms and the multiple groups of sample single-dimensional impurity spectrograms. Specifically, a layer separation model is constructed based on a plurality of sample composite impurity spectrograms and a plurality of groups of sample single-dimensional impurity spectrograms, characteristics of the composite impurity spectrograms, such as peak positions, peak shapes, peak intensities and spectral curve trends, are extracted by using an image analysis algorithm and a chemical metering principle, different impurity spectral information in the composite spectrograms is separated into different layers by adopting methods of cluster analysis, principal component analysis and the like, the sample single-dimensional impurity spectrograms are used as reference standard verification optimization, and M kinds of impurity spectral information and below in the composite impurity spectrograms can be accurately separated into corresponding layers by the model through repeated adjustment of model parameters and algorithm structures, so that clear independent spectrum data sources are provided for impurity content identification.

And step S240, constructing and obtaining M impurity content identification branches based on the M sample impurity spectrum atlas and the M sample impurity content identification sets. Specifically, M impurity content identification branches are constructed based on M sample impurity spectrum atlas and M sample impurity content identification sets. Branches are identified for each impurity content, with the corresponding sample impurity spectral atlas as the primary source of learning data. Regression analysis algorithms in machine learning, such as linear regression, multiple regression, or support vector regression, etc., are employed. And taking the spectral characteristic data in the sample impurity spectrogram set as an input variable, and taking the impurity content data in the sample impurity content identification set corresponding to the spectral characteristic data as an output variable to perform model training. In the training process, parameters of the model, such as regression coefficients, kernel function parameters and the like, are continuously optimized, and the fitting degree of the model on the relation between the spectral characteristics and the impurity content is improved. Through training and verification of a large amount of data, each impurity content identification branch can accurately predict the impurity content according to the specific impurity spectral characteristics of the input unknown sample, so that the content identification function of M sample impurity types is realized.

And step S250, connecting the M impurity content identification branches in parallel, and after marking the M impurity content identification branches by adopting the M sample impurity types, connecting the output end of the layer separation model with the input end of the M impurity content identification branches to complete the construction of the impurity detection model. Specifically, M impurity content recognition branches are connected in parallel, so that they can independently perform content recognition work on respective corresponding impurity types. And then respectively identifying the M impurity content identification branches by adopting M sample impurity types, and determining the function and the corresponding impurity type of each branch. And connecting the output end of the layer separation model with the input ends of M impurity content identification branches to construct a complete impurity detection model. When the spectrum data of an unknown sample is input, firstly, a layer separation model is entered, the impurity spectrum information in the unknown sample is separated into different layers according to different impurity types, then, the spectrum data of each layer is respectively input into corresponding impurity content identification branches, the branches are used for predicting and analyzing the impurity content, and finally, the content information of M sample impurity types in the unknown sample is obtained, so that the comprehensive and accurate detection of the sample impurities is realized.

And step S260, storing the M sample impurity types and M impurity content identification branches in an associated mode, and completing construction of the detection model call table. Specifically, M sample impurity types and M impurity content identification branches are stored in an associated mode, and a detection model call table is constructed. In this process, M sample impurity types are used as index keys, and corresponding M impurity content identification branches are used as storage contents or link addresses pointing to the branches. For example, the storage may be in the form of a database table in which one column records the sample impurity type and the other column records the impurity content identification branch related information corresponding thereto. Therefore, when impurity detection is actually carried out, when a certain specific sample impurity type needs to be detected, the corresponding impurity content identification branch can be rapidly positioned only by inquiring the detection model calling table, and therefore the branch is called to carry out impurity content detection. The method greatly improves the efficiency and the flexibility of the impurity detection model in practical application, and ensures that the impurities in the glucosamine production process can be detected and monitored rapidly and accurately.

And step S300, carrying out impurity accumulation rule analysis on the plurality of quality control samples, and positioning K groups of key impurity types of K key steps based on analysis results, wherein the K key steps have K execution sequence identifiers. Specifically, first, for a plurality of liquid quality control samples, an impurity detection model is used, a first quality control sample is divided into N step sub-samples, N real-time impurity spectrograms are generated by an infrared spectrometer, the N real-time impurity spectrograms are synchronized to a layer separation model to separate H real-time single-dimensional spectrograms of H sample impurity types, H real-time impurity content is obtained by inputting impurity content identification branches according to corresponding relations to form first real-time impurity information, N real-time impurity information is obtained by processing the N real-time impurity spectrograms according to the first real-time impurity information, a plurality of sample impurity content sets are obtained by polymerization extraction very worth of the first sample impurity content set, the sample impurity content sets are repeatedly operated, an impurity distribution matrix is constructed by taking the sample impurity type sets as constraints, K key steps and K key impurity types are located and K execution sequence identifiers are defined through operations such as accumulation calculation, accumulation rate analysis, cross-step impurity removal contribution analysis and single step weighting calculation.

In one possible implementation manner, the impurity accumulation rule analysis is performed on the plurality of quality control samples, and K groups of K key impurity types of K key steps are located based on the analysis result, wherein the K key steps have K execution sequence identifications, and step S300 further includes step S310, using the impurity detection model, polling the plurality of quality control samples to obtain a plurality of sample impurity content sets. Specifically, after finishing quality control liquid samples collected in each step of glucosamine production, starting an impurity detection model to carry out polling detection, dividing one sample into N step sub-samples according to production key elements, generating N real-time impurity spectrograms containing impurity molecular information by using a high-precision infrared spectrometer, sending the first real-time impurity spectrograms to a layer separation model, separating H real-time single-dimensional spectrograms of H sample impurity types according to a spectral feature library and an image analysis algorithm, introducing M impurity content identification branches according to a corresponding relation, obtaining H real-time impurity contents to form first real-time impurity information by means of regression analysis, processing the N real-time impurity spectrograms to obtain N real-time impurity information, summarizing and integrating, extracting extremum information to construct a first sample impurity content set, and repeating operation to finish detection of all samples to obtain a plurality of sample impurity content sets.

And step S320, constructing and generating an impurity distribution matrix according to the plurality of sample impurity content sets by taking the sample impurity type set as a constraint. Specifically, an impurity distribution matrix is constructed with a set of sample impurity types as constraints, each row representing a specific impurity type such as heavy metal impurities, solvent residues, oxidation byproducts, and each column corresponding to each production step from raw material pretreatment to final product formation. The matrix element value is derived from a plurality of sample impurity content sets, namely detection concentration or quality data of certain impurity in a specific production step, for example, the value of a heavy metal impurity in a corresponding column of a decalcification step is the average concentration or total quality of the step in all samples, and the construction mode integrates dispersion data, so that the impurity distribution is clear, and a data base is built for analyzing and accumulating rules.

And step S330, carrying out accumulation rule analysis according to the impurity distribution matrix, and positioning the K key steps and K groups of key impurity types based on analysis results. The method comprises the steps of firstly calculating the accumulation amount of impurities based on an impurity distribution matrix, accumulating the element values of each production step column according to the impurity types of each row of the matrix, for example, obtaining the data and accumulated total amount of heavy metal impurities under multiple steps to know the overall accumulation condition and screen key impurity types, then carrying out accumulation rate analysis, calculating the impurity content change rate of adjacent steps to determine the growth rate, for example, decalcification to deacidification, changing the heavy metal impurities to find out the key production interval, further carrying out cross-step impurity removal contribution analysis, evaluating the impurity removal effect and the subsequent influence of a specific step, for example, the effect of a purification step on solvent residues, and determining the improvement of impurity removal efficiency of a key hinge step.

And step S340, obtaining K sample impurity content sets and the K execution sequence identifications from the plurality of execution sequence identifications and the plurality of sample impurity content sets according to the K key steps. Specifically, finally, carrying out single-step weighted calculation by comprehensively considering factors such as impurity accumulation amount, accumulation rate, cross-step impurity removal contribution and the like, giving weight to impurity types and production steps according to the influence degree of each factor on product quality and production process, giving high weight to accumulation amount, high rate and low impurity removal contribution, accurately positioning K key steps and K groups of key impurity types from a plurality of steps and impurity types through weighted calculation, if a production step with high accumulation amount, high rate and low impurity removal contribution of specific impurities is the key step, the corresponding impurity is the key impurity type, simultaneously extracting and executing sequence identification according to the K key steps, calling the corresponding sample impurity content set, and providing target and data support for process optimization and impurity control.

In a possible implementation manner, the impurity detection model is used to poll and detect the plurality of quality control samples to obtain a plurality of sample impurity content sets, step S310 further includes step S311, and after the first quality control sample is divided into N step sub-samples, an infrared spectrometer is used to generate N real-time impurity spectrograms of the N step sub-samples. Specifically, first, for a first quality control sample, it is precisely divided into N step sub-samples according to each key link or step feature in the production process. Since each step has a corresponding liquid quality control sample, this division can carefully reflect the variation of the product and impurities at different stages of production. After the division is completed, the sub-samples of the N steps are respectively detected by using a high-precision infrared spectrometer. The infrared spectrometer generates N real-time impurity spectrograms of N step sub-samples by emitting infrared light in a specific wavelength range and detecting the absorption condition of the sample on the light. These spectra are presented in the form of unique spectral curves which contain rich information such as the vibration absorption peak position, intensity, etc. of specific chemical bonds in the impurity molecules, which can be used as important basis for identifying the type and content of the impurities.

Step S312, synchronizing the first real-time impurity spectrogram to the layer separation model of the impurity detection model, and performing layer separation based on the impurity types to obtain H real-time single-dimensional spectrograms of H sample impurity types, wherein H is a positive integer less than or equal to M. Specifically, the first real-time impurity spectrogram is transmitted to a layer separation model in a pre-constructed impurity detection model. The model relies on a powerful algorithm and a rich impurity spectrum feature library, and performs fine layer separation operation on the composite first real-time impurity spectrogram based on impurity types. In the separation process, the model decomposes the first real-time impurity spectrogram into H real-time single-dimensional spectrograms of H sample impurity types according to characteristic differences, such as characteristic absorption peak positions, peak shapes, peak relative intensities and the like, of different impurities, which are shown on the spectrograms. The H is a positive integer less than or equal to M, which means that the number of the separated impurity types does not exceed the total number M of the impurity types of the sample, thereby ensuring the accuracy and the effectiveness of the separation result, each single-dimensional spectrogram focuses on a specific impurity type, and laying a solid foundation for the accurate subsequent impurity content identification.

Step S313, according to the correspondence between the impurity types of the H samples and the M impurity content identification branches in the impurity detection model, guiding the H real-time single-dimensional spectrograms to input the H impurity content identification branches for impurity content directional identification, so as to obtain H real-time impurity contents, where the H real-time impurity contents form first real-time impurity information. Specifically, according to the preset corresponding relation between the impurity types of the H samples and M impurity content identification branches in the impurity detection model, the H real-time single-dimensional spectrograms are guided and input into the corresponding H impurity content identification branches accurately. These impurity content identification branches are intelligent modules trained based on a large number of sample data, and they employ advanced machine learning algorithms, such as linear regression, multiple regression, or neural network regression, etc. After the single-dimensional spectrogram is input, the branch model can extract key characteristic data in the spectrogram, and according to mathematical relations between the data and samples with known impurity contents, directional identification calculation of the impurity contents is performed, and finally H real-time impurity contents are obtained. The real-time impurity contents together form first real-time impurity information, which can reflect the impurity content condition of the first quality control sample in the current production step in real time, and provide timely and accurate data support for quality monitoring in the production process.

Step S314, and the like, analyzing the N real-time impurity spectrograms by adopting the impurity detection model to obtain N real-time impurity information. Specifically, according to the same flow and method, the N real-time impurity spectrograms are sequentially and deeply analyzed by using the impurity detection model. For each real-time impurity spectrogram, repeating the operation steps of layer separation, single-dimensional spectrogram acquisition, impurity content directional identification and the like, thereby obtaining N pieces of complete real-time impurity information. The real-time impurity information covers various information such as impurity types, content and related spectral characteristics of the first quality control sample in each subdivision production step, and the information is comprehensive and detailed description of the impurity condition of the first quality control sample, so that abundant data resources are provided for subsequent further analysis and processing.

And step S315, after the N pieces of real-time impurity information are polymerized based on the impurity types, extracting an impurity content extremum to obtain a first sample impurity content set. Specifically, after N pieces of real-time impurity information are acquired, the information is subjected to an aggregation operation based on the impurity type. The real-time impurity information of the same impurity type in different production steps is integrated to form an information sequence about each impurity type in the whole production process of the first quality control sample. Then, an impurity content extremum is extracted from these information sequences. For example, the maximum content value and the minimum content value of a certain impurity in all production steps are found out, and the key information such as the corresponding production steps is found out. In this way, the plurality of real-time impurity information is concentrated into a more representative and instructive first sample impurity content set. The content set not only contains content range information of various impurity types in the first quality control sample, but also indirectly reflects the variation trend and key control points of the impurities in the production process, and provides important reference basis for optimizing the production process and controlling the impurities.

Step S316, and the like, using the impurity detection model, polling the plurality of quality control samples to obtain the impurity content sets of the plurality of samples. Specifically, taking the detection and analysis flow of the first quality control sample as an example, and so on, the impurity detection model is adopted to carry out comprehensive polling detection on all the plurality of quality control samples. The above-described set of operational flows from sub-sample division, spectral diagram generation, layer separation, impurity content identification to impurity content set generation are repeated for each quality control sample. After the detection of all quality control samples is completed, the impurity content sets of the samples obtained by all the samples are summarized and integrated, and finally, a plurality of sample impurity content sets which can reflect the impurity conditions in different production steps in the whole glucosamine production process are successfully obtained. The data sets provide comprehensive and detailed data bases for the subsequent deep analysis of impurity accumulation rules, positioning key steps and key impurity types in the production process, making a targeted impurity removal strategy and other works, and are key points for realizing the efficient quality control and process optimization of the glucosamine production process.

In a possible implementation manner, the accumulation rule analysis is performed according to the impurity distribution matrix, and the K key steps and the K groups of key impurity types are located based on the analysis result, and step S330 further includes step S331, in which an accumulation amount calculation is performed according to the impurity distribution matrix, so as to obtain M impurity accumulation amount sequences corresponding to the M sample impurity types. Specifically, first, an impurity accumulation amount calculation work is performed in accordance with the constructed impurity distribution matrix. Each row of the impurity distribution matrix represents a sample impurity type, each column corresponds to a production step, and the matrix element value is the detection concentration or quality of the impurity in the step. For each of the M sample impurity types, the element values (i.e., impurity concentration or quality data) in the respective columns of production steps are cumulatively summed along their corresponding rows. For example, for a first sample impurity type, the impurity content of the first production step is added to the impurity content of the second production step in sequence until all production steps are added, thus resulting in an aggregate total of the impurity types. According to the method, M sample impurity types are calculated one by one, and M impurity accumulation sequences corresponding to the M sample impurity types are finally obtained, wherein the sequences clearly show the overall accumulation condition of each impurity in the whole production flow.

And S332, carrying out accumulation rate analysis according to the impurity distribution matrix to obtain M accumulation rate sequences. Specifically, after the impurity accumulation amount sequence is obtained, accumulation rate analysis is performed in accordance with the impurity distribution matrix. For each sample impurity type, the rate of change of impurity content between adjacent production steps was calculated by dividing the concentration change of the impurity before and after the step by the step execution time. For example, for the second sample impurity type, the accumulation rate from the third production step to the fourth production step is calculated, expressed by (impurity concentration of the fourth production step-impurity concentration of the third production step)/the execution period between the two steps. According to such a calculation, calculation is performed between each adjacent production steps for each sample impurity type, thereby obtaining a sequence including the concentration change rates of the plurality of steps, and finally obtaining M accumulation rate sequences. The sequences can reflect the growth speed of the impurities in different production stages, and help to find out the key interval of abnormal impurity growth.

And step S333, performing cross-step impurity removal contribution analysis according to the M impurity accumulation amount sequences to obtain M impurity removal contribution rate sequences. Specifically, cross-step impurity removal contribution analysis was performed based on the existing M impurity accumulation sequences. For each sample impurity type, the effect of removing the impurity in a particular production step and the effect of such removal on the impurity content in subsequent production steps were examined. For example, for the third sample impurity type, the amount removed in the purification step and the degree of reduction in the impurity content in the concentration step after removal were analyzed. And quantifying the contribution value of each step to specific impurity removal by comparing the change conditions of the impurity content in different production steps, so as to obtain M impurity removal contribution rate sequences. The sequences can determine which production steps play a key role in removing specific impurities in the whole impurity removing process, and provide important basis for the optimization of the subsequent process.

And step S334, performing single-step weighted calculation on the M impurity accumulation amount sequences, the M accumulation rate sequences and the M impurity removal contribution rate sequences to obtain M impurity removal priority sequences. Specifically, single-step weighting calculation is performed on M impurity accumulation amount sequences, M accumulation rate sequences, and M impurity removal contribution rate sequences. The weight is determined according to the importance degree of each factor on the product quality and the production process. For example, higher weights are given at the corresponding production steps for impurity types that have high accumulation, fast accumulation rate, and low contribution to cross-step impurity removal. For each production step and each impurity type, comprehensively considering the data information in the three sequences, and calculating according to a set weight calculation method to obtain a numerical value comprehensively reflecting the impurity type impurity removal priority in the production step. Through calculation of all impurity types and production steps, M impurity removal priority sequences are finally obtained, and the sequences can more comprehensively and accurately measure the impurity removal importance degree of each production step for different impurity types.

And step S335, carrying out highest-level calling from the M impurity removal priority sequences to obtain M impurity removal priority steps. Specifically, the highest-level call is performed from the M impurity removal priority sequences, that is, the production step corresponding to the maximum value in each impurity removal priority sequence is found. The production steps are the steps with the highest impurity removal priority for the corresponding impurity types, and M impurity removal priority steps are obtained by extracting the steps. These steps have the most critical role in controlling specific impurity types throughout the production process, and are important concerns for subsequent focus process optimization and impurity control.

Step S336, the M impurity removal priority steps are polymerized, and the K key steps are obtained. Specifically, the M impurity removal priority steps are subjected to polymerization treatment, and possibly repeated steps are removed, so that K key steps are obtained. These key steps cover the production links that are most critical for the control of various impurity types throughout the production process.

And step S337, aggregating the M sample impurity types according to the mapping relation between the M impurity removal priority steps and the K key steps to obtain the K groups of key impurity types. Specifically, according to the mapping relation between the M impurity removal priority steps and the K key steps, sample impurity types associated with the K key steps are found, and the impurity types are aggregated to obtain K groups of key impurity types. The determination of the key steps and the key impurity types provides a definite target and direction for further carrying out targeted process optimization and impurity control on the glucosamine production process, and is beneficial to improving the product quality and the production efficiency.

And step S400, taking the K groups of key impurity types as impurity removal guide, carrying out coupling optimization on the K key steps according to the K execution sequence identifiers, and outputting K optimized process parameters. Specifically, K groups of key impurity types are used as impurity removal guide, all key step information is determined according to K execution sequence identifications, target impurity concentrations of M sample impurity types are obtained interactively for each key step and corresponding impurity types, impurity concentration constraints of the key steps are determined according to the target impurity concentrations, initial process parameters are obtained interactively, a sample impurity concentration set, a sample process parameter set and a sample concentration change set are collected in a constraint networking mode by using the key steps and the impurity types as constraints, an impurity prediction model is constructed through multiple regression analysis, the initial process parameters are subjected to solution expansion by using a preset parameter adjustment scale to obtain updated process parameters, the prediction model is used for combining the impurity concentration constraints to screen out K groups of reliable process parameters, K groups of reliable process parameters are obtained through one-by-one optimization of K key steps, and finally K optimized process parameters are output through coupling optimization such as permutation and combination enumeration, simulation production, cost analysis and the like.

In a possible implementation manner, the K groups of key impurity types are used as impurity removal guidance, the K key steps are coupled and optimized according to the K execution sequence identifications, K optimized process parameters are output, and step S400 further includes step S410, and M target impurity concentrations of the M sample impurity types are obtained interactively. Specifically, first, M target impurity concentrations for M sample impurity types are obtained by interacting with a specialized database, an industry standard data interface, or a large number of experimental data records accumulated internally. These target impurity concentrations are determined based on intensive studies of the quality requirements of glucosamine products and on purity standards commonly accepted in the industry. For example, for certain metal ion impurities, the target impurity concentration may be set to an extremely low level of parts per million to ensure the safety and effectiveness of the glucosamine product in pharmaceutical or other high-end applications. The data provides a clear impurity control direction for process parameter optimization of the subsequent key steps, and is an important reference of the whole optimization process.

Step S420, respectively extracting a first key step and a first group of key impurity types from the K key steps and the K groups of key impurity types according to the K execution orders. Specifically, according to the K execution sequence identifications, the first key step and the first group of key impurity types are accurately extracted from the determined K key steps and K groups of key impurity types. Taking the glucosamine production flow as an example, if K key steps are distributed in links of raw material pretreatment, reaction synthesis, intermediate product purification and the like, then the first key step determined according to the execution sequence identification may be a specific step in the reaction synthesis link, and the first group of key impurity types corresponding to the specific step may be byproduct impurities easily generated in the reaction process or residual impurities difficult to remove in the raw material. The step defines a specific object which is required to be subjected to process parameter optimization at present, so that the subsequent data acquisition and model construction are more targeted.

Step S430, obtaining a first impurity concentration constraint from the M target impurity concentration calls according to the first set of critical impurity types. Specifically, from the first set of critical impurity types, concentration data matching the M target impurity concentrations is recalled, thereby obtaining a first impurity concentration constraint. This constraint defines the range of concentrations allowed for the first set of critical impurity types in the first critical step. For example, if the first set of critical impurity types contains two impurities, then the first impurity concentration constraint would dictate the respective highest allowable concentration values for the two impurities at the end of the first critical step. The method provides strict limiting conditions for the adjustment of subsequent process parameters, ensures that the control of impurity concentration is not ignored due to excessive pursuit of other indexes when the process parameters are optimized, and ensures that the product quality is always in a controllable range.

Step S440, interactively obtaining the first initial process parameters of the first key step. Specifically, the first initial process parameters of the first critical step are obtained by interacting with the production facility control system, the process parameter record document, or the operator's empirical data. These initial process parameters cover the various process conditions involved in this step, such as reaction temperature, reaction pressure, reaction time, reactant concentration ratios, etc. For example, where the first critical step is a chemical reaction step, the initial process parameters may include a reaction temperature set at 50 degrees celsius, a reaction pressure of 2 atmospheres, a reaction time of 3 hours, and a concentration ratio of reactant a to reactant B of 2:1, etc. These data reflect the actual operating parameter settings of the first critical step in the current production process and are the starting point for process parameter optimization.

And S450, carrying out networking data acquisition by using the first key step and the first group of key impurity types as data calling constraints, and obtaining a first sample impurity concentration set, a first sample process parameter set and a first sample concentration change set. Specifically, the first key step and the first group of key impurity types are used for data call constraint, and the data call constraint is accessed to a wide chemical industry data network, a scientific research institution shared data platform, a historical production data warehouse in an enterprise and the like to perform comprehensive networking data acquisition. In this process, a first sample impurity concentration set is obtained, the data set comprising actual impurity concentration measurements of a first set of critical impurity types during a plurality of first critical step-like processes, the first sample process parameter set collecting various process parameter combinations employed during the like processes, the first sample concentration change set recording concentration changes of the first set of critical impurity types as process parameters change. For example, from experimental data of a plurality of different production lots but similar to the first critical step, concentration variation data of the first set of critical impurity types at different combinations of reaction temperatures, pressures and times are extracted, which will provide sufficient material for constructing accurate impurity prediction models.

Step S460, performing multiple regression analysis on the first sample impurity concentration set, the first sample process parameter set and the first sample concentration change set, and fitting to generate a first impurity prediction model. Specifically, an in-depth multiple regression analysis is performed on the collected first sample impurity concentration set, first sample process parameter set, and first sample concentration variation set. And fitting to generate a first impurity prediction model by using advanced statistical analysis software and a data modeling algorithm. The model establishes a mathematical relationship between process parameters and impurity concentrations, for example, it can predict concentration values of a first set of critical impurity types at the end of a first critical step based on the entered process parameters such as reaction temperature, pressure, and time. The model is a core tool for subsequent evaluation and optimization of the process parameters, can rapidly evaluate the influence of different process parameter combinations on the impurity concentration under the condition of not carrying out actual production test, and greatly improves the efficiency of process parameter optimization.

Step S470, performing the de-expansion of the first initial process parameters by adopting a preset parameter adjustment scale to obtain a first group of updated process parameters. Specifically, a preset parameter adjustment scale is adopted to carry out de-expansion on the first initial process parameters, so that a first group of updated process parameters are obtained. The preset tuning scale is determined based on understanding of the production process, industry experience, and mastering of the equipment performance. For example, if the initial reaction temperature is 50 degrees celsius, and the preset parameter adjustment scale specifies the temperature adjustment range to be ±10 degrees celsius, the reaction temperature in the updated process parameters may take values of 40 degrees celsius, 45 degrees celsius, 55 degrees celsius, 60 degrees celsius, and the like. In this way, the limitations of the initial process parameters are broken, a wider process parameter space is explored, and the possibility of finding a better process parameter combination is increased.

And step S480, evaluating the first group of updated process parameters by adopting the first impurity prediction model to obtain a first group of reliable process parameters. Specifically, a first set of updated process parameters is evaluated using a first impurity prediction model. Inputting the first set of updated process parameters into the first impurity prediction model to obtain a first set of predicted concentration variation sets comprising concentration variation conditions of the first set of critical impurity types predicted under different combinations of updated process parameters. And then, traversing and comparing the first group of predicted concentration change sets by adopting the first impurity concentration constraint, and screening a first group of reliable process parameters of which the impurity concentration meets the first impurity concentration constraint from the first group of updated process parameters. For example, if a first impurity concentration constraint specifies that the concentration of an impurity should not exceed 0.1%, then only those updated process parameter combinations that result in an impurity concentration below 0.1% will be selected as a first set of reliable process parameters in the first set of predicted concentration variation sets. This step ensures that the optimized process parameters can meet the impurity concentration control requirements and find the optimal solution in a wider parameter range.

Step S490, and so on, taking the K groups of key impurity types as impurity removal guide, carrying out process parameter optimization on the K key steps according to the K execution sequence identifications, and outputting K groups of reliable process parameters. Specifically, according to the steps, K groups of key impurity types are used as impurity removal guide, and the K key steps are sequentially subjected to process parameter optimization according to K execution sequence identifiers, so that K groups of reliable process parameters are finally output.

Step S4100, coupling optimization is performed on the K groups of reliable process parameters, and the K optimized process parameters are output. Specifically, after the independent optimization of each key step is completed, the K sets of reliable process parameters are optimized for coupling. Considering that the key steps are not isolated, but are mutually influenced and restrained, for example, the process parameter change of the previous key step may influence the raw material quality or reaction condition of the next key step, so that the process parameters of the key steps need to be comprehensively coordinated through methods such as permutation and combination enumeration, simulated production, cost analysis and the like. For example, in the simulation production process, different K sets of reliable process parameter combinations are tried, indexes such as reagent consumption, energy consumption, impurity removal period, production efficiency and the like under each combination are calculated, and finally, a set of optimal process parameter combinations are determined, namely K optimal process parameters are output, and the optimal process parameters can achieve optimal production benefit while the effective control of impurities is realized on the whole.

In one possible implementation, the first set of updated process parameters is evaluated using the first impurity prediction model to obtain a first set of reliable process parameters, and step S480 further includes step S481 of searching for a first set of key impurity concentrations from the impurity distribution matrix using the first key step and the first set of key impurity types as search constraints. Specifically, the first key step and the first group of key impurity types are used as search constraint conditions, and accurate search is performed in the constructed impurity distribution matrix. Each row of the impurity distribution matrix represents an impurity type, and each column corresponds to a production step in which information on the detected concentration or quality of each impurity at the corresponding step is stored. Since the first critical step and the first set of critical impurity types are determined, the corresponding cell data can be quickly located through the row-column index, thereby obtaining the first set of critical impurity concentrations. For example, if the first critical step is a "mid-reaction" step in the glucosamine production process, the first set of critical impurity types includes "side reaction product a" and "residual feedstock impurity B", then the data in the intersecting cells of the "mid-reaction" column and "side reaction product a" row, "residual feedstock impurity B" row are found in the impurity distribution matrix, and these data are the first set of critical impurity concentrations. The concentration data reflect the actual concentration levels of the key impurities in the key step in the past production practice, and provide important reference basis for subsequent model evaluation.

And step S482, synchronizing the first group of key impurity concentrations and the first group of updated process parameters to the first impurity prediction model evaluation to obtain a first group of predicted concentration change sets. Specifically, the obtained first group of key impurity concentrations and the first group of updated process parameters are synchronously input into a first impurity prediction model for evaluation. The first impurity prediction model is constructed by multiple regression analysis based on a plurality of sample data, which establishes a mathematical relationship between process parameters and impurity concentrations. After the first group of key impurity concentrations and the first group of updating process parameters are input, the model carries out complex operation according to an internal algorithm and a data relation, and when the updating process parameters are adopted, the change condition of the first group of key impurity concentrations is predicted, so that a first group of predicted concentration change set is obtained. For example, if the parameters such as reaction temperature, pressure and reaction time are adjusted in the first set of updated process parameters, the model calculates the values of the concentration changes of the "side reaction product a" and the "residual raw material impurity B" under the new parameters based on these parameter changes and the correlation with the impurity concentration, and these values constitute a first set of predicted concentration change sets that exhibit the possible trend of the change in impurity concentration assuming updated process parameters are employed.

Step S483, selecting the first set of reliable process parameters with impurity concentration satisfying the first impurity concentration constraint from the first set of updated process parameters by using the first impurity concentration constraint to traverse and compare the first set of predicted concentration change sets. Specifically, a first set of reliable process parameters meeting the requirements are screened from a first set of updated process parameters by adopting first impurity concentration constraint traversal comparison of a first set of predicted concentration change sets. The first impurity concentration constraint defines the range of concentrations allowed for the first set of critical impurity types in the first critical step, which is set based on product quality criteria and manufacturing process requirements. In the comparison process, whether each predicted concentration value in the first group of predicted concentration change sets meets the first impurity concentration constraint is checked one by one. For example, if the first impurity concentration constraint specifies that the concentration of "side reaction product a" should not exceed 0.5% and the concentration of "residual feedstock impurity B" should not exceed 0.3%, then the corresponding first set of updated process parameters are deemed satisfactory as a first set of reliable process parameters only if the predicted concentration values of the corresponding impurities in the first set of predicted concentration variation sets are below these defined values. The reliable process parameters can theoretically ensure that the concentration of key impurities is controlled in a reasonable range, embody the verification of the effectiveness of updating and adjusting the process parameters, and provide feasible parameter selection for the optimization of the subsequent production process.

In a possible implementation manner, the K sets of reliable process parameters are optimized in a coupling manner, the K optimized process parameters are output, step S4100 further includes step S4101, and the K sets of reliable process parameters are enumerated in a permutation and combination manner to obtain a plurality of sets of coupling process parameters. Specifically, firstly, the K groups of reliable process parameters are arranged and enumerated, and as each group covers multiple key step optimization parameters, the overall arrangement and combination can generate a large number of multiple groups of coupling process parameters, for example, 27 groups can be generated when 3 key steps are included and 3 reliable parameters are valued respectively, the mode can systematically explore the matching effect of different step parameters, the better combination is mined, the situation that the global optimum is missed due to the fact that the local optimum is trapped is avoided, and the firm foundation is built for screening the optimal process parameter combination.

Step S4102, performing glucosamine production simulation on the plurality of groups of coupling process parameters to obtain a plurality of pieces of process performance information, wherein the process performance information comprises reagent consumption, impurity removal period, production efficiency and production energy consumption. Specifically, for a plurality of groups of coupling process parameters, chemical production simulation software or a built mathematical model is utilized to simulate the glucosamine production process, after parameters are input, conditions of each link are recorded in detail to obtain process performance information, such as reagent consumption, impurity removal period, production efficiency and production energy consumption data under different combinations, wherein the reagent consumption is determined according to reaction consumption, the impurity removal period is calculated according to simulation steps and removal rates, the production efficiency is embodied in unit time yield, the production energy consumption is obtained by integrating various operation consumption, and the information can reflect the actual expression of the coupling process parameters and provide quantitative data support for cost analysis.

Step S4103, obtaining a target coupling process parameter combination by performing cost analysis on the plurality of process performance information, wherein the target coupling process parameter combination includes the K optimized process parameters. Specifically, by cost analyzing the plurality of process performance information, cost factors of various aspects are comprehensively weighed. In the aspect of reagent dosage, the reagent cost directly influences the production cost, the larger the dosage is, the higher the cost is, the long impurity removal period possibly means long equipment occupation time and increased labor cost, the lower production efficiency can lead to the increase of the fixed cost of unit product allocation, and the higher the production energy consumption can lead to the increase of the energy cost. Corresponding cost weights are set according to these factors, for example, for enterprises where energy costs are relatively high, the weight of production energy consumption may be relatively large. And then calculating the total cost corresponding to each group of coupling process parameters, and screening out one or more groups of coupling process parameter combinations with the lowest cost by comparing the total cost, wherein the coupling process parameter combinations are target coupling process parameter combinations, and the target coupling process parameter combinations comprise K finally determined optimized process parameters. Under the premise of ensuring the quality of the product (through effective control of the types of key impurities), the optimized process parameters realize the minimization of the production cost, and improve the economic benefit and the competitiveness of the glucosamine production process.

And S500, carrying out online detection function configuration on the K key steps according to the K groups of key impurity types and a detection model call table to obtain K online detection modules. Specifically, K impurity content identification branches matched with K groups of key impurity types are screened out from a detection model call table according to the K groups of key impurity types, K online detection modules are built for the K key steps by integrating and configuring the branches and providing hardware equipment such as a high-precision spectrometer, a sample acquisition and transmission device, a data processing unit and the like, each module corresponds to a specific key step and can accurately detect the content of the key impurity types, then the modules are calibrated and optimized by utilizing standard impurity samples with known concentration, spectrometer parameters are adjusted, an algorithm model is optimized to ensure detection precision and stability, meanwhile, sample acquisition and transmission efficiency and data processing speed are improved, so that the K online detection modules can effectively perform impurity deviation dynamic detection early warning on the K key steps in the production of glucosamine, and the stable and consistent product quality is ensured.

And step S600, after the process parameters of the K key steps are updated by adopting the K optimized process parameters, carrying out impurity deviation dynamic detection and early warning on the glucosamine production process by the K online detection modules. Specifically, when glucosamine is produced, the process parameters of K key steps are updated according to K optimized process parameters, after production is started, K online detection modules acquire samples according to preset conditions by means of a special sample acquisition device for each key step and send the samples to a high-precision spectrometer to acquire spectrum data, a data processing unit inputs the samples to corresponding impurity content identification branches to calculate key impurity content, and then the key impurity content is compared with a preset standard content range to calculate an offset value, if the offset value is in a normal range, the production is continued, if the offset value exceeds the normal range, the modules immediately trigger early warning signals, the early warning signals are transmitted to monitoring staff in various modes and attach detailed offset information, and the monitoring staff evaluate the production process according to the offset and take adjustment measures such as checking process parameters or equipment conditions to correct the offset, so that stable, efficient and standard production is ensured.

The embodiment of the application adopts quality control samples of a plurality of steps obtained by sampling, extracts an impurity type set by combining a production log, and configures an impurity detection model and a call table. And analyzing the accumulation rule of the sample impurities, positioning key steps and the impurity types corresponding to the key steps, optimizing process parameters based on the key steps, and configuring an online detection module. The optimized process parameters are applied in the key steps, and the dynamic monitoring and early warning of impurity deviation are realized through the online detection module, so that the technical effects of improving the accuracy and efficiency of impurity control in the glucosamine production and ensuring the consistency and stability of the product quality are achieved.

Hereinabove, the impurity detection method in the glucosamine production process according to the embodiment of the present invention is described in detail with reference to fig. 1. Next, an impurity detection system in a glucosamine production process according to an embodiment of the present invention will be described with reference to fig. 2.

The impurity detection system in the glucosamine production process is used for solving the technical problem of insufficient impurity control in the existing glucosamine production process, and achieves the technical effects of improving the accuracy and efficiency of impurity control in the glucosamine production and ensuring the consistency and stability of the product quality. The impurity detection system in the glucosamine production process comprises a quality control sample acquisition module 10, an impurity detection model configuration module 20, an accumulation rule analysis module 30, a coupling optimization module 40, an online detection function configuration module 50 and a dynamic detection early warning module 60.

The quality control sample acquisition module 10 is configured to obtain a plurality of quality control samples for a plurality of production steps by sampling during a glucosamine test production process, wherein the plurality of production steps have a plurality of execution order identifiers.

The impurity detection model configuration module 20 is configured to obtain a sample impurity type set from the interactive production log, and then configure an impurity detection model according to the sample impurity type set, so as to obtain a detection model call table.

The accumulation law analysis module 30 is configured to perform impurity accumulation law analysis on the plurality of quality control samples, and locate K groups of K key impurity types of K key steps based on analysis results, where the K key steps have K execution order identifiers.

The coupling optimization module 40 is configured to perform coupling optimization on the K key steps according to the K execution order identifiers by using the K key impurity types as a impurity removal guide, and output K optimized process parameters.

The online detection function configuration module 50 is configured to perform online detection function configuration on the K key steps according to the K sets of key impurity types and the detection model call table, so as to obtain K online detection modules.

The dynamic detection and early warning module 60 is used for carrying out dynamic detection and early warning on impurity deviation in the glucosamine production process through the K online detection modules after the process parameters of the K key steps are updated by adopting the K optimized process parameters.

Next, the specific configuration of the impurity detection model configuration module 20 will be described in detail. As described above, after the interactive production log obtains the sample impurity type set, performing impurity detection model configuration according to the sample impurity type set to obtain a detection model call table, the impurity detection model configuration module 20 further includes a sample impurity spectrum atlas obtaining unit, configured to perform networking data call according to M sample impurity types in the sample impurity type set, obtain M sample impurity spectrum atlas of the M sample impurity types, where the M sample impurity spectrum atlas has M sample impurity content identification sets; the system comprises a networking data calling unit, a layer separation model constructing unit, an identification branch constructing unit and an end connecting unit, wherein the networking data calling unit is used for carrying out networking data calling by taking M sample impurity types as constraints to obtain a plurality of sample composite impurity spectrograms and a plurality of groups of sample single-dimensional impurity spectrograms, the layer separation model constructing unit is used for constructing and obtaining a layer separation model based on the plurality of sample composite impurity spectrograms and the plurality of groups of sample single-dimensional impurity spectrograms, the identification branch constructing unit is used for constructing and obtaining M impurity content identification branches based on the M sample impurity spectrograms and M sample impurity content identification sets, and the end connecting unit is used for connecting the M impurity content identification branches in parallel and connecting the output end of the layer separation model with the input end of the M impurity content identification branches after the M sample impurity types are adopted to identify the M impurity content identification branches so as to complete the construction of the impurity detection model;

and the detection model call table construction unit is used for storing the M sample impurity types and M impurity content identification branches in a correlated mode to finish construction of the detection model call table.

Next, the specific configuration of the accumulation law analysis module 30 will be described in detail. As described above, the impurity accumulation rule analysis module 30 further includes a sample impurity content set acquisition unit configured to poll the plurality of quality control samples to obtain a plurality of sample impurity content sets using the impurity detection model, an impurity distribution matrix generation unit configured to construct and generate an impurity distribution matrix based on the plurality of sample impurity content sets with the sample impurity type set as a constraint, an impurity type positioning unit configured to perform accumulation rule analysis based on the impurity distribution matrix and position the K key steps and the K group key impurity types based on the analysis result, and a content set calling unit configured to call K sample impurity content sets and the K execution identifiers from the plurality of execution order identifiers and the plurality of sample impurity content sets according to the K key steps.

The sample impurity content set acquisition unit further comprises a real-time impurity spectrogram generation subunit, a layer separation subunit, a directional identification subunit, a first extreme value detection subunit and a real-time sample impurity type acquisition subunit, wherein the real-time impurity spectrogram generation subunit is used for dividing a first quality control sample into N step subsamples and then generating N real-time impurity spectrograms of the N step subsamples by adopting an infrared spectrometer, the layer separation subunit is used for synchronizing a first real-time impurity spectrogram to the layer separation model of the impurity detection model, carrying out layer separation based on an impurity type to obtain H real-time single-dimensional spectrograms of H sample impurity types, H is a positive integer smaller than or equal to M, the directional identification subunit is used for guiding the H real-time single-dimensional impurity spectrograms to be input into the impurity content identification branches for carrying out content directional identification according to the corresponding relation between the H sample impurity types and the M impurity content identification branches in the impurity detection model, the H real-time single-dimensional impurity spectrograms are used for carrying out content directional identification, the H real-time impurity content detection subunit is used for carrying out layer separation based on the first real-time impurity spectrogram, the extreme value detection unit is used for obtaining extreme value information by adopting the N sample impurity type acquisition subunit, the real-time impurity detection subunit is used for carrying out extreme value detection by carrying out extreme value detection on the N impurity type acquisition unit, and adopting the impurity detection model to poll and detect the plurality of quality control samples to obtain impurity content sets of the plurality of samples.

The impurity type positioning unit further comprises an accumulation amount calculation subunit, an accumulation rate analysis subunit, an impurity removal contribution analysis subunit, a weighting calculation subunit, a highest-level calling subunit and a highest-level calling subunit, wherein the accumulation amount calculation subunit is used for carrying out accumulation rate analysis according to the impurity distribution matrix to obtain M accumulation rate sequences, the impurity removal contribution analysis subunit is used for carrying out cross-step impurity removal contribution analysis according to the M impurity accumulation rate sequences to obtain M impurity removal contribution rate sequences, the weighting calculation subunit is used for carrying out single-step weighting calculation on the M impurity accumulation rate sequences, the M accumulation rate sequences and the M impurity removal contribution rate sequences to obtain M impurity removal priority sequences, the highest-level calling subunit is used for carrying out highest-level calling from the M impurity removal priority sequences to obtain M impurity removal priority sequences, the highest-level calling subunit is used for obtaining M impurity removal priority sequences, and the M impurity removal priority sequences are used for obtaining the key impurity types according to the M impurity accumulation rate sequences, and the key impurity types are obtained by the key impurity type positioning unit.

Next, a specific configuration of the coupling optimization module 40 will be described in detail. As described above, with the K groups of critical impurity types as impurity removal guide, performing coupling optimization on the K critical steps according to the K execution sequence identifiers, and outputting K optimized process parameters, the coupling optimization module 40 further includes an interaction acquisition unit, configured to interactively obtain M target impurity concentrations of the M sample impurity types; the system comprises a K key step and K group key impurity types, an impurity concentration constraint acquisition unit, a parameter step acquisition unit, a networking data acquisition unit, a multiple regression analysis unit, a solution expansion unit and a process parameter updating unit, wherein the K key step and the K group key impurity types are respectively extracted from the K key step and the K group key impurity types according to the K execution sequences, the impurity concentration constraint acquisition unit is used for acquiring a first impurity concentration constraint from the M target impurity concentration calls according to the first group key impurity types, the parameter step acquisition unit is used for interactively acquiring a first initial process parameter of the first key step, the networking data acquisition unit is used for carrying out networking data acquisition by using the first key step and the first group key impurity types as data call constraints to acquire a first sample impurity concentration set, a first sample process parameter set and a first sample concentration change set, the multiple regression analysis unit is used for carrying out multiple regression analysis on the first sample impurity concentration set, the first sample concentration set and the first sample concentration change set to generate a first impurity prediction model through fitting, the solution expansion unit is used for carrying out the updating of the first parameter updating unit by presetting the initial parameter updating the first parameter of the initial parameter updating unit, the parameter evaluation unit is used for evaluating the first group of updated process parameters by adopting the first impurity prediction model to obtain a first group of reliable process parameters, the process parameter optimization unit is used for performing process parameter optimization on the K key steps according to the K execution sequence identifications by analogy with the K key impurity types as impurity removal guidance to output K groups of reliable process parameters, and the coupling optimization unit is used for performing coupling optimization on the K groups of reliable process parameters to output the K optimized process parameters.

The parameter evaluation unit further comprises a search constraint subunit, a parameter synchronization subunit and a constraint traversal subunit, wherein the search constraint subunit is used for searching and obtaining a first group of key impurity concentration from the impurity distribution matrix by taking the first key step and the first group of key impurity type as search constraints, the parameter synchronization subunit is used for synchronizing the first group of key impurity concentration and the first group of updating process parameters to the first impurity prediction model to evaluate so as to obtain a first group of predicted concentration change set, and the constraint traversal subunit is used for obtaining the first group of reliable process parameters with the impurity concentration meeting the first impurity concentration constraint from the first group of updating process parameters by adopting the first impurity concentration constraint traversal to compare the first group of predicted concentration change set.

The coupling optimization unit further comprises an arrangement and combination enumeration subunit, a production simulation subunit and a cost analysis subunit, wherein the arrangement and combination enumeration subunit is used for carrying out arrangement and combination enumeration on the K groups of reliable process parameters to obtain a plurality of groups of coupling process parameters, the production simulation subunit is used for carrying out glucosamine production simulation on the plurality of groups of coupling process parameters to obtain a plurality of pieces of process performance information, the process performance information comprises reagent consumption, impurity removal period, production efficiency and production energy consumption, and the cost analysis subunit is used for obtaining a target coupling process parameter combination by carrying out cost analysis on the plurality of pieces of process performance information, and the target coupling process parameter combination comprises the K pieces of optimization process parameters.

The impurity detection system in the glucosamine production process provided by the embodiment of the invention can execute the impurity detection method in the glucosamine production process provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

Although the present application makes various references to certain modules in a system according to an embodiment of the present application, any number of different modules may be used and run on a user terminal and/or a server, and each unit and module included are merely divided according to functional logic, but are not limited to the above-described division, so long as the corresponding functions can be implemented, and in addition, specific names of each functional unit are only for convenience of distinguishing from each other, and are not intended to limit the scope of protection of the present application.

The above embodiments do not limit the scope of the present application. It will be apparent to those skilled in the art that various modifications, combinations, and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the scope of the present application. In some cases, the acts or steps recited in the present application may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. The method for detecting impurities in the glucosamine production process is characterized by comprising the following steps:

Obtaining a plurality of quality control samples for a plurality of production steps by sampling during a glucosamine test production process, wherein the plurality of production steps have a plurality of execution order identifiers;

After the interactive production log obtains a sample impurity type set, carrying out impurity detection model configuration according to the sample impurity type set to obtain a detection model call table;

Performing impurity accumulation rule analysis on the plurality of quality control samples, and positioning K groups of key impurity types of K key steps based on analysis results, wherein the K key steps have K execution sequence identifiers;

taking the K groups of key impurity types as impurity removal guide, carrying out coupling optimization on the K key steps according to the K execution sequence identifications, and outputting K optimized process parameters;

Performing online detection function configuration on the K key steps according to the K groups of key impurity types and a detection model call table to obtain K online detection modules;

and after the K optimized process parameters are adopted to update the process parameters of the K key steps, carrying out impurity deviation dynamic detection and early warning on the glucosamine production process through the K online detection modules.

2. The method for detecting impurities in a glucosamine production process as set forth in claim 1, wherein after obtaining a sample impurity type set from an interactive production log, performing an impurity detection model configuration based on the sample impurity type set to obtain a detection model call table, the method comprising:

according to M sample impurity types in the sample impurity type set, carrying out networking data call to obtain M sample impurity spectrum atlas of the M sample impurity types, wherein the M sample impurity spectrum atlas is provided with M sample impurity content identification sets;

The M sample impurity types are used as constraints to carry out networking data calling, so that a plurality of sample compound impurity spectrograms and a plurality of groups of sample single-dimensional impurity spectrograms are obtained;

constructing and obtaining a layer separation model based on the multiple sample composite impurity spectrograms and the multiple groups of sample single-dimensional impurity spectrograms;

M impurity content identification branches are constructed and obtained based on the M sample impurity spectrum atlas and the M sample impurity content identification sets;

The M impurity content identification branches are connected in parallel, after the M impurity content identification branches are identified by adopting the M sample impurity types, the output end of the layer separation model is connected with the input end of the M impurity content identification branches, and the construction of the impurity detection model is completed;

and storing the M sample impurity types and M impurity content identification branches in an associated mode, and completing construction of the detection model call table.

3. The method for detecting impurities in a glucosamine production process as set forth in claim 2, wherein an impurity accumulation law analysis is performed on said plurality of quality control samples, and K sets of K key impurity types for K key steps are located based on the analysis results, said method comprising:

Adopting the impurity detection model to poll and detect the plurality of quality control samples to obtain a plurality of sample impurity content sets;

Constructing and generating an impurity distribution matrix according to the impurity content sets of the plurality of samples by taking the sample impurity type set as constraint;

Analyzing an accumulation rule according to the impurity distribution matrix, and positioning the K key steps and K groups of key impurity types based on analysis results;

and obtaining K sample impurity content sets and the K execution sequence identifications from the plurality of execution sequence identifications and the plurality of sample impurity content sets according to the K key steps.

4. The method for detecting impurities in a glucosamine manufacturing process as set forth in claim 3, wherein said plurality of quality control samples are polled for detection using said impurity detection model to obtain a plurality of sample impurity content sets, said method comprising:

Dividing a first quality control sample into N step sub-samples, and generating N real-time impurity spectrograms of the N step sub-samples by using an infrared spectrometer;

Synchronizing a first real-time impurity spectrogram to the layer separation model of the impurity detection model, and carrying out layer separation based on impurity types to obtain H real-time single-dimensional spectrograms of H sample impurity types, wherein H is a positive integer less than or equal to M;

according to the corresponding relation between the impurity types of the H samples and the M impurity content identification branches in the impurity detection model, guiding the H real-time single-dimensional spectrograms to be input into the H impurity content identification branches for carrying out impurity content directional identification, and obtaining H real-time impurity contents, wherein the H real-time impurity contents form first real-time impurity information;

similarly, analyzing the N real-time impurity spectrograms by adopting the impurity detection model to obtain N real-time impurity information;

The N real-time impurity information is polymerized based on the impurity type, and then impurity content extremum extraction is carried out to obtain a first sample impurity content set;

and similarly, using the impurity detection model, polling the plurality of quality control samples to obtain the plurality of sample impurity content sets.

5. The method for detecting impurities in a glucosamine production process as set forth in claim 3, wherein an analysis of a cumulative rule is performed based on the impurity distribution matrix, and wherein the K key steps and K groups of key impurity types are located based on the analysis result, the method comprising:

Performing cumulative amount calculation according to the impurity distribution matrix to obtain M impurity cumulative amount sequences corresponding to the M sample impurity types;

Performing accumulation rate analysis according to the impurity distribution matrix to obtain M accumulation rate sequences;

Performing cross-step impurity removal contribution analysis according to the M impurity accumulation sequences to obtain M impurity removal contribution rate sequences;

carrying out single-step weighted calculation on the M impurity accumulation amount sequences, the M accumulation rate sequences and the M impurity removal contribution rate sequences to obtain M impurity removal priority sequences;

Carrying out highest-level calling from the M impurity removal priority sequences to obtain M impurity removal priority steps;

the M impurity removal priority steps are polymerized to obtain K key steps;

and aggregating the M sample impurity types according to the mapping relation between the M impurity removal priority steps and the K key steps to obtain the K groups of key impurity types.

6. The method for detecting impurities in a glucosamine production process as set forth in claim 3, wherein said K sets of critical impurity types are used as impurity removal guides, said K critical steps are coupled and optimized according to said K execution sequence identifications, and K optimized process parameters are outputted, said method comprising:

interactively obtaining M target impurity concentrations of the M sample impurity types;

Respectively extracting a first key step and a first group of key impurity types from the K key steps and the K groups of key impurity types according to the K execution sequence identifications;

obtaining a first impurity concentration constraint from the M target impurity concentration calls according to the first set of key impurity types;

interactively obtaining first initial process parameters of the first key step;

performing networking data acquisition by using the first key step and the first group of key impurity types as data calling constraints to obtain a first sample impurity concentration set, a first sample process parameter set and a first sample concentration change set;

Performing multiple regression analysis on the first sample impurity concentration set, the first sample process parameter set and the first sample concentration change set, and fitting to generate a first impurity prediction model;

performing the de-expansion of the first initial process parameters by adopting a preset parameter adjustment scale to obtain a first group of updated process parameters;

Evaluating the first set of updated process parameters by adopting the first impurity prediction model to obtain a first set of reliable process parameters;

And by analogy, carrying out process parameter optimization on the K key steps according to the K execution sequence identifications by taking the K groups of key impurity types as impurity removal guide, and outputting K groups of reliable process parameters;

and performing coupling optimization on the K groups of reliable process parameters, and outputting the K optimized process parameters.

7. The method for detecting impurities in a glucosamine manufacturing process as set forth in claim 6, wherein said first set of updated process parameters is evaluated using said first impurity predictive model to obtain a first set of reliable process parameters, said method comprising:

searching from the impurity distribution matrix to obtain a first set of key impurity concentrations by taking the first key step and the first set of key impurity types as search constraints;

Synchronizing the first group of key impurity concentrations and the first group of updated process parameters to the first impurity prediction model for evaluation to obtain a first group of predicted concentration change sets;

And obtaining the first set of reliable process parameters with impurity concentration meeting the first impurity concentration constraint from the first set of updated process parameters by traversing and comparing the first set of predicted concentration change sets by adopting the first impurity concentration constraint.

8. The method for detecting impurities in a glucosamine production process as set forth in claim 6, wherein said K sets of reliable process parameters are coupled and optimized for outputting said K optimized process parameters, said method comprising:

performing permutation and combination enumeration on the K groups of reliable process parameters to obtain a plurality of groups of coupling process parameters;

Performing glucosamine production simulation on the multiple groups of coupling process parameters to obtain multiple pieces of process performance information, wherein the process performance information comprises reagent consumption, impurity removal period, production efficiency and production energy consumption;

And obtaining a target coupling process parameter combination by carrying out cost analysis on the plurality of process performance information, wherein the target coupling process parameter combination comprises the K optimized process parameters.

9. A system for detecting impurities in a glucosamine production process, wherein the system is adapted to perform the method for detecting impurities in a glucosamine production process as set forth in any one of claims 1-8, the system comprising:

The quality control sample acquisition module is used for acquiring a plurality of quality control samples of a plurality of production steps by sampling in the glucosamine test production process, wherein the plurality of production steps have a plurality of execution sequence identifications;

the impurity detection model configuration module is used for carrying out impurity detection model configuration according to the sample impurity type set after the sample impurity type set is obtained by the interactive production log, so as to obtain a detection model call table;

the accumulation rule analysis module is used for carrying out impurity accumulation rule analysis on the plurality of quality control samples and positioning K groups of key impurity types of K key steps based on analysis results, wherein the K key steps have K execution sequence identifications;

The coupling optimization module is used for carrying out coupling optimization on the K key steps according to the K execution sequence identifications by taking the K groups of key impurity types as impurity removal guide, and outputting K optimized process parameters;

The online detection function configuration module is used for carrying out online detection function configuration on the K key steps according to the K groups of key impurity types and the detection model call table to obtain K online detection modules;

And the dynamic detection and early warning module is used for carrying out dynamic detection and early warning on impurity deviation in the glucosamine production process through the K online detection modules after the K optimized process parameters are adopted to update the process parameters of the K key steps.