Chemical instrument following analysis system

Abstract

The present application relates to a kind of chemical instrument follow-up analysis system, including instrument model construction subsystem, experimental data analysis subsystem and instrument model correction subsystem;According to the type of analysis instrument, an instrument model is constructed, and the corresponding correlation index is established with instrument characteristic data and analysis type data to form the neural network of model, on the one hand, the experimental project information can be analyzed using instrument model, the corresponding adaptable and suggested analysis instrument is analyzed further, and the information of corresponding analysis instrument is presented, on the other hand, according to actual experimental result, instrument characteristic data in instrument model is corrected, instrument model is updated, and instrument characteristic data can be updated according to experimental result to improve the precision of system to provide accurate instrument report information, to assist experimenter to further judge.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of chemical instrument management, and more particularly, to a chemical instrument following analysis system. BACKGROUND

[0002] Instrument analysis refers to a class of methods that use relatively complex or special instruments and equipment to obtain information about the chemical composition, content of components, and chemical structure of a substance by measuring certain physical or physical-chemical properties of the substance and changes in the properties.

[0003] The analysis objects of instrument analysis are generally semi-micro (0.01-0.1 g), micro (0.1-10 mg), and ultra-micro (<0.1 mg) components, and the sensitivity is high; while the analysis objects of chemical analysis are generally semi-micro (0.01-0.1 g) and macro (>0.1 g) components, and the accuracy is high.

[0004] Instrument analysis can be roughly divided into: electrochemical analysis method, nuclear magnetic resonance spectroscopy, atomic emission spectroscopy, gas chromatography, atomic absorption spectroscopy, high-performance liquid chromatography, ultraviolet-visible spectroscopy, mass spectrometry, infrared spectroscopy, and other instrument analysis methods.

[0005] Instrument analysis is an analysis method that uses experimental phenomena that can directly or indirectly represent the characteristics (such as physical, chemical, and physiological properties) of a substance to transform information about the composition, content, distribution, or structure of the substance into information that can be directly perceived by humans through a probe or sensor, an amplifier, an analysis converter, and the like. In other words, instrument analysis is an analysis method that uses the basic principles of various disciplines and advanced technologies such as electricity, optics, precision instrument manufacturing, vacuum, and computers to explore the chemical characteristics of a substance. Therefore, instrument analysis is a highly integrated branch of science and technology that embodies the cross-disciplinary nature of science and technology. Instrument analysis has developed very rapidly and has a very broad application prospect.

[0006] And instrument analysis due to more accurate and less error, so widely used, but there are still some problems, due to the independence of instrument analysis data, due to different chemical instruments, different principles, but the instrument in long time use will exist certain error, and due to the use of the wrong way may lead to experimental results appear deviation, at the same time due to a kind of material analysis may be through a variety of different chemical instruments for analysis, and the analysis results need to be artificial for judgment and integration, undoubtedly lead to the reliability of data decline, the experimenter needs to analyze or combine more chemical instruments in order to get more comprehensive analysis results, but due to the high cost of chemical instruments, so the effective analysis information can not be obtained, leading to the whole analysis project into the block. SUMMARY

[0007] Therefore, the present application aims to provide a chemical instrument following analysis system.

[0008] In order to solve the above technical problems, the technical scheme of the present application is:

[0009] A chemical instrument following analysis system, comprising an instrument model construction subsystem, an experimental data analysis subsystem and an instrument model correction subsystem;

[0010] The instrument model construction subsystem is used to generate an instrument model, the instrument model comprising a plurality of modular instrument information, the modular instrument information comprising instrument content data, instrument characteristic data and analysis type data, the instrument content data reflecting the type of chemical instrument, the instrument characteristic data reflecting the analysis characteristics of the chemical instrument, and the analysis type data reflecting the analysis situation applicable to the chemical instrument, the instrument model construction subsystem comprising a sample input module, a characteristic network construction module and an analysis network construction module, the sample input module comprising different chemical instrument information input according to different types of chemical instruments, and processing the input chemical instrument information in a fixed format to generate modular information, the characteristic network construction module being configured with a preset feature extraction strategy and a plurality of feature association conditions, the instrument characteristics of the instrument characteristic data being extracted by the feature extraction strategy, and the correlation between the instrument characteristics in different modular instrument information being determined by the feature association conditions to generate a feature association index, and the analysis network construction module being configured with a preset type extraction strategy and a plurality of type association conditions, the analysis type of the analysis type data being extracted by the type extraction strategy, and the analysis type index being generated according to the modular instrument information having the same analysis type;

[0011] The experimental data analysis subsystem includes an information acquisition module, an information index module, and an information analysis module. The information acquisition module is configured to acquire experimental project information, which includes experimental content data and experimental requirement data. The information index module extracts corresponding instrument characteristics according to the experimental content data in the experimental project information, determines corresponding analysis types according to the experimental requirement data, and determines corresponding modular instrument information from the instrument model according to the instrument characteristics and the analysis types. The information analysis module filters the modular instrument information according to the experimental requirement data and outputs instrument report information according to the obtained modular instrument information.

[0012] The instrument model correction subsystem includes a correction trigger module and an information correction module. The correction trigger module is configured with a correction trigger condition. When the experimental project information meets the correction trigger condition, the corresponding experimental content data is extracted. The information correction module generates new instrument characteristic data according to the experimental content data and updates the instrument model. Meanwhile, the feature network construction module updates the corresponding feature association index according to the new instrument characteristic data.

[0013] Further, the system further includes an abnormality analysis subsystem. The chemical instrument information includes standard result information and abnormal result information. The standard result information reflects the analysis results of the instrument in a normal state, and the abnormal result information reflects the analysis results of the instrument in an abnormal state. The sample input module compares the standard result information and the abnormal result information to generate abnormal feature items. The instrument characteristic data is generated according to the abnormal feature items. The abnormal feature items include abnormal feature content and abnormal feature types.

[0014] The abnormality analysis subsystem includes an abnormality recognition module and an abnormality analysis module. The abnormality recognition module determines whether there is the same abnormal feature content in the modular information according to the detection results in the experimental project information. If there is the same abnormal feature content, the abnormal feature type is sent to the abnormality analysis module. The abnormality analysis module retrieves corresponding modular instrument information according to the abnormal feature content and outputs abnormal report information according to the obtained modular instrument information and the abnormal feature type.

[0015] Further, the sample input module further includes a behavior analysis unit. The behavior analysis unit is connected to an analysis collection end. The analysis collection end is configured to collect experimental behavior information. The experimental behavior information reflects the behavior of the user during the experiment. The behavior analysis unit analyzes the experimental behavior information to generate the abnormal feature type.

[0016] Further, the abnormality analysis subsystem further includes a data compensation module. The data compensation module generates a compensation strategy according to the abnormal feature type and corrects the detection results through the compensation strategy.

[0017] Further, the analysis acquisition end is arranged as an image acquisition device, the image acquisition device is arranged to face the experiment space of the chemical instrument, and the behavior analysis unit generates the abnormal feature type according to the obtained image information.

[0018] Further, the behavior analysis unit is configured with a behavior analysis strategy, the behavior analysis strategy analyzes standard behavior features to form an image verification sequence, the image verification sequence includes a plurality of image verification features, and whether the corresponding image verification feature exists in the image information is sequentially compared, if not, the image verification feature is marked, and the abnormal feature type is generated according to the marked image verification feature.

[0019] Further, the standard behavior feature is generated according to experiment project information.

[0020] Further, the standard behavior feature is generated according to instrument feature data.

[0021] Further, the analysis acquisition end is a state monitoring unit, the state monitoring unit is connected to the corresponding chemical instrument and monitors the working state of the chemical instrument to generate the abnormal feature type.

[0022] The technical effects of the present application mainly embody in the following aspects: by such arrangement, an instrument model is constructed according to the type of the chemical instrument, and corresponding association indexes are established by instrument feature data and analysis type data to form a neural network of the model, on the one hand, the instrument model can be used to analyze the experiment project information, obtain corresponding adaptable and recommended chemical instruments for further analysis, and present information of the corresponding chemical instruments, on the other hand, the instrument feature data in the instrument model is corrected according to the actual experiment result, the instrument model is updated, the instrument feature data can be updated according to the experiment result to improve the precision of the system to provide accurate instrument report information, and assist the experimenter to make further judgment. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 The present application is a chemical instrument following analysis system system architecture principle diagram;

[0024] Figure 2 The present application is an instrument model construction subsystem principle diagram;

[0025] Figure 3 The present application is an experiment data analysis subsystem principle diagram;

[0026] Figure 4 The present application is an instrument model correction subsystem principle diagram;

[0027] Figure 5 The present application is an abnormal analysis subsystem principle diagram.

[0028] Reference signs: 100, instrument model construction subsystem; 110, sample input module; 120, feature network construction module; 130, analysis network construction module; 200, experimental data analysis subsystem; 210, information acquisition module; 220, information index module; 230, information analysis module; 300, instrument model correction subsystem; 310, correction triggering module; 320, information correction module; 400, anomaly analysis subsystem; 410, anomaly identification module; 420, anomaly analysis module; 430, data compensation module. DETAILED DESCRIPTION

[0029] The specific embodiments of the present application are further described in detail below with reference to the accompanying drawings, so that the technical scheme of the present application is easier to understand and master.

[0030] A chemical instrument following analysis system comprises an instrument model construction subsystem 100, an experimental data analysis subsystem 200, and an instrument model correction subsystem 300;

[0031] The instrument model construction subsystem 100 is used to generate an instrument model, which includes several modular instrument information, including instrument content data, instrument feature data, and analysis type data. First, the instrument content data reflects the type of chemical instrument, which includes at least the following two data, the purpose of which is to accurately determine the type of instrument, 1. Instrument classification, such as conductometric analyzer, potentiometric titration analyzer, optical spectrum analyzer, solid-state nuclear magnetic resonance analyzer, biological nuclear magnetic resonance analyzer, gas-solid chromatographic analyzer, gas-liquid chromatographic analyzer, etc. 2. Instrument model, mainly to distinguish the same / different instruments under the same classification. The above two pieces of information constitute the instrument content data. Second, the instrument feature data is the core content of the present application and needs to be expanded. The instrument feature data reflects the analysis characteristics of the chemical instrument, specifically including the following contents, 1. The purpose of the instrument analysis, such as material composition analysis, solution saturation analysis, solution ionization degree analysis, and element composition analysis of the substance, etc. 2. The analysis characteristics of the instrument, including the range of substances suitable for the instrument, the accuracy of the instrument analysis, and the type of substances suitable for the instrument. 3. The analysis result characteristics of the instrument, such as abnormal output waveform caused by improper operation, or abnormal output results caused by physical component damage or installation problems of the instrument. The abnormal result information and normal result information can be generated, which will be described in detail below. Finally, the analysis type data reflects the analysis situation suitable for the chemical instrument, including the following contents: 1. The environmental conditions suitable for the instrument, such as some instruments that need to be operated in vacuum or sterile conditions, or have certain requirements for the environment, temperature and humidity. 2. The physical characteristics of the substances suitable for the instrument, such as the volume of the substances suitable for the instrument and the form of the substances suitable for the instrument. Through the above three characteristics, the information of the instrument can be collected in different dimensions.

[0032] The instrument model construction subsystem 100 includes a sample input module 110, a feature network construction module 120, and an analysis network construction module 130. The sample input module 110 inputs different chemical instrument information according to different types of chemical instruments and processes the input chemical instrument information in a fixed format to generate modular information. First, the sample input module 110 inputs different chemical instrument information according to different types of chemical instruments. The input methods are as follows: 1. manual input; 2. detection by a collection device; and 3. obtaining information recorded in an external database. The three methods can also be combined, for example, the model of a chemical instrument can be manually input, and the corresponding instrument content data can be automatically generated through network information, such as the classification of the instrument, the analysis features of the instrument, the analysis purpose of the instrument, and the environmental conditions suitable for the instrument, the physical characteristics of the substances used, and the like. At the same time, experimental data and experimental results obtained using the information can be obtained through feedback detection, thereby generating instrument analysis result features. The corresponding chemical instrument information can be generated. The key part of the present application is to uniformly format the chemical instrument information to generate corresponding modular instrument information. A major difficulty here is that the same substance or instrument can be described differently, with multiple different or similar textual descriptions, which makes it difficult to format uniformly. The sample input module 110 is configured by the following means. The sample input module 110 includes a storage database for storing different data. The data is stored in the form of text. At the same time, a keyword database is used to format and judge the content in the text through keyword association. The keyword database is constructed and input in real time by the backend. At the same time, a construction output inquiry unit is set up, a similarity threshold is set, and texts with the same chemical instrument information and a text overlap higher than the similarity threshold are extracted and output to the display end for efficient data unification. The construction of the keyword database in the later stage is more abundant, the information granularity is more precise, and the reliability of the data is improved. As a preferred embodiment, the text information processed uniformly above is established with an encoding index, which can specifically be 8 bits of instrument content data, 128 bits of instrument feature data, and 16 bits of analysis type data.

[0033] The feature network construction module 120 is configured with a preset feature extraction strategy and a plurality of feature association conditions. The instrument features of the instrument feature data are extracted by the feature extraction strategy, and the association relationship of the instrument features in different modular instrument information is determined by the feature association conditions to generate a feature association index. The feature network construction module 120 mainly has two functions. The first function is to extract the instrument features in the instrument feature data, that is, the feature extraction strategy. The feature extraction strategy includes the following steps: information classification step, feature recognition step, and feature extraction step. First, the information classification step classifies the information according to different contents in the instrument feature data. The feature recognition step retrieves resource information of a feature recognition database. The resource information of the feature recognition database is indexed by information classification. Then, the feature recognition conditions are generated according to the retrieved resource information to identify the instrument features from the instrument feature data. The feature extraction step extracts the instrument features according to the identified data. The other function is to establish the association of the instrument features to form a feature association index. The feature association index data includes the associated target address and the associated content. The associated target address reflects the addresses of two modular instrument information and the addresses of the corresponding instrument feature data. The associated content is preferably divided into the following parts: different dimensions and advantages. For example, the instrument types of A and B are the same, but the precision of the B instrument is higher than that of the A instrument. Therefore, the A instrument and the B instrument establish a feature association index for the instrument features of the precision value, and the associated content is high precision. The types of the A and B instruments are different, but both the A and B instruments can detect the same chemical substances by different detection methods, such as one by spectral analysis and one by electrochemical analysis. Therefore, the different analysis methods can be used as the associated content in the associated content part. In this way, a feature network can be constructed based on the relationship of the instrument features.

[0034] The analysis network construction module 130 is configured with a preset type extraction strategy and a plurality of type association conditions. The analysis type of the analysis type data is extracted by the type extraction strategy, and the analysis type index is generated according to the modular instrument information with the same analysis type. The type extraction strategy of the analysis network construction module 130 is the same as the feature extraction strategy, and will not be described in detail. The key is the generation of the analysis type index. For example, the analysis type index includes the different application conditions of the same instrument and the different application environments of the same instrument, such as industrial chemical instruments and laboratory chemical instruments. The division provides a basis for subsequent information analysis.

[0035] The experimental data analysis subsystem 200 includes an information acquisition module 210, an information indexing module 220, and an information analysis module 230. The information acquisition module 210 is used to acquire experimental project information, which includes experimental content data and experimental requirement data. First, the information acquisition module 210 acquires experimental project information. Experimental project information is recorded in the form of text at the beginning of an experiment, and experimental data is recorded during the experiment. The experimental data constitutes the experimental content data, and the experimental requirement data is generated based on the experimental project to determine the experimental requirements of the user. The experimental requirement data can also be generated by the user inputting the data. In this way, the experimental project information can be acquired.

[0036] The information indexing module 220 extracts corresponding instrument characteristics from the experimental content data in the experimental project information and determines the corresponding analysis type based on the experimental requirement data. Based on the instrument characteristics and the analysis type, the corresponding modular instrument information is determined from the instrument model. Based on the experimental content data, the matching corresponding related instrument characteristics can be obtained. Then, based on the experimental requirement data, the analysis type is filtered to determine the corresponding modular information. Specifically, the experimental purpose, the analyzed substance, the analyzed result, and other characteristics can be generated based on the experimental content data. The chemical instrument used is positioned to the baseline modular instrument information. Based on other experimental content data, the corresponding characteristic correlation index can be positioned. Based on the experimental requirement data, the analysis type index can be positioned. Through these two indexes, other modular information can be determined. Generally, the analysis type index is determined by exclusion. After positioning a number of modular instrument information using the characteristic correlation index, the analysis type index is used for analysis. The data that does not meet the experimental requirements is excluded to obtain the modular information.

[0037] The information analysis module 230 filters the modular instrument information based on the experimental requirement data and outputs instrument report information based on the obtained modular instrument information. Finally, based on the experimental requirement data and the modular information, the instrument report information can be obtained. For example, the detection precision is insufficient, for example, A chemical instrument can be selected. For example, the detection dimension needs to be rich, for example, B chemical instrument can be selected. The problem is analyzed based on the experimental requirement and result data, and reasonable suggestions in the instrument analysis direction are given.

[0038] The instrument model correction subsystem 300 comprises a correction trigger module 310 and an information correction module 320. The correction trigger module 310 is configured with a correction trigger condition. When the experimental item information meets the correction trigger condition, the corresponding experimental content data is extracted. First, the correction trigger condition, for example, a new experimental result is generated, or the actual experimental result and the experimental result corresponding to the sample have a large difference. Here, the comparison content is involved. Because the sample data may be manually entered or obtained through an external network, the information accuracy is low. By obtaining the actual experimental content data and correcting the corresponding data through more actual samples, the influence of the sample on the result tends to be consistent, thereby obtaining accurate instrument characteristic data. As for the correction variable, a correction value is assigned according to the deviation result, so that the correction is dynamically and gradually convergent, ensuring the adaptability and continuity of the correction. The information correction module 320 generates new instrument characteristic data according to the experimental content data and updates the instrument model. Meanwhile, the characteristic network construction module 120 updates the corresponding characteristic correlation index according to the new instrument characteristic data. After the data information is corrected, the characteristic correlation index is updated to ensure the consistency of the data.

[0039] The abnormality analysis subsystem 400, the chemical instrument information includes standard result information and abnormal result information, the standard result information reflects the analysis result of the instrument in the normal state, the abnormal result information reflects the analysis result of the instrument in the abnormal state, the sample input module compares the standard result information and the abnormal result information to generate an abnormal feature item, the instrument feature data is generated according to the abnormal feature item, and the abnormal feature item includes abnormal feature content and abnormal feature type; Because the chemical instrument information includes experimental results and experimental processes, by collecting these information, whether the operation error occurs in the instrument use process or whether the instrument itself has a problem can be judged, and the above problems are difficult to find, so that the experimental project obtains the wrong result. First, a large number of sample inputs can obtain the analysis result in the normal experimental state, including column change graph, waveform change graph, and can also obtain the analysis result in the abnormal state, so that the difference between the analysis result in the normal state and the analysis result in the abnormal state can be analyzed, and the corresponding abnormal feature item can be obtained. For example, if a chemical instrument is not sealed during use, the test result obtained is that the working time of the vacuum pump is longer than the working time of the vacuum pump in the normal state, at this time, the time difference between the two can be used as a feature item, and the feature content is unsealed, and the abnormal feature type is that the working time of the vacuum pump exceeds the preset value. The standard behavior feature is generated according to the experimental project information. The standard behavior feature is generated according to the instrument feature data. The standard behavior feature can be generated according to the experimental project information, and the general experimental project information will record the standard operation behavior, and the corresponding standard behavior feature can be obtained by intercepting the text, and the corresponding standard behavior feature can also be obtained by the instrument feature data of the corresponding modular instrument information in the instrument model.

[0040] The abnormality analysis subsystem 400 comprises an abnormality identification module 410 and an abnormality analysis module 420; the abnormality identification module 410 judges whether there is the same abnormal feature content from the modular information according to the detection result in the experimental project information, if there is the same abnormal feature content, the abnormal feature type is sent to the abnormality analysis module 420, and the abnormality identification module 410 of the abnormality analysis subsystem 400 is used for identification, the corresponding abnormal feature content is obtained from the experimental project information, for example, the working time of the vacuum pump exceeds the preset value, if the content appears, the corresponding modular instrument information is called through the abnormality analysis module 420. The abnormality analysis module 420 calls the corresponding modular instrument information according to the abnormal feature content and outputs the abnormal report information according to the obtained modular instrument information and the abnormal feature type. At this time, all related information of the chemical instrument is called, and the related information such as temperature, humidity, use steps and the like is called through the corresponding abnormal feature type, and then the abnormal report information is output. The user can discover the instrument abnormality or improper operation in the first time, and the problem is avoided. The advantage of the present application is that the previous abnormality identification needs to be combined with human experience, so the problem is generally difficult to be discovered. The instrument model is established in the present application, the corresponding instrument features are associated, the data is continuously collected through the test to enrich the sample, the effect of accurate analysis is achieved, and the analysis report is made for the subsequent experimental results. For example, a plurality of different experimental results reflect a same change, the experimental results are all judged as extreme situation experimental results, that is, the abnormal data is discarded, and the present application can extract the commonality of the experimental results, judge whether there is instrument abnormality or improper operation, and remind the user to adjust in the first time.

[0041] The key part of the application, the sample input module 110 also includes a behavior analysis unit, the behavior analysis unit is connected to the analysis collection end, the analysis collection end is used to collect experimental behavior information, the experimental behavior information reflects the behavior of the user in the experimental process, the behavior analysis unit analyzes the experimental behavior information to generate the abnormal feature type. Through the behavior analysis unit, the user's experimental behavior information is analyzed to make a judgment on the user's behavior. Specifically, the analysis collection end is set as an image collector, the image collector is arranged to face the experimental space of the chemical instrument, and the behavior analysis unit generates the abnormal feature type according to the obtained image information. The behavior analysis unit is configured with a behavior analysis strategy, the behavior analysis strategy analyzes standard behavior characteristics to form an image verification sequence, the image verification sequence includes a plurality of image verification characteristics, and the image information is sequentially compared to determine whether the corresponding image verification characteristics exist, if not, the image verification characteristics are marked and the abnormal feature type is generated according to the marked image verification characteristics. First, the behavior analysis is more complex, and the behavior analysis strategy is described in detail. First, the image collector is installed, and the image collector is arranged to face the experimental space to ensure that the key part of the experimental instrument (at least including the shape and pattern characteristics capable of image recognition and positioning, and the part capable of monitoring the operation position of the instrument) in the collection space. The first step is to calibrate the image characteristics, and first, the three-dimensional characteristics of the chemical instrument are input, and the system obtains the three-dimensional characteristics of the chemical instrument. At the same time, the image collector calibrates the actual image characteristics, determines the corresponding device image characteristics, and then calibrates the key part of the user. Preferably, a calibration sticker is established on the experimenter's hand, and further image characteristics of auxiliary instruments are established. The behavior analysis strategy divides the information in the picture according to the operation required by the user during the instrument analysis, for example, the user will first add A solution to the container, and then add B solution to the container. Then the image verification sequence generates two instructions to verify whether A solution is added and whether B solution is added. Through image recognition technology, it can be judged whether the above-mentioned solution is added, because the device and auxiliary instrument (solution bottle) have been calibrated at the analysis collection end, so the corresponding image verification characteristics can be verified to mark whether there is an abnormal feature type. Then, based on this abnormal feature type, it can be judged whether the user has abnormal behavior during operation. The analysis collection end is a state monitoring unit, which is connected to the corresponding chemical instrument and monitors the working state of the chemical instrument to generate the abnormal feature type. Through the state monitoring unit directly connected to the chemical instrument, the working state information is obtained to determine whether the instrument is abnormal, such as working temperature, working current, etc.

[0042] The abnormality analysis subsystem 400 further comprises a data compensation module 430, which generates a compensation strategy according to the abnormality feature type and corrects the detection result through the compensation strategy. Since the detection result can be corrected through the compensation strategy if the corresponding abnormal behavior is determined, the result can be used as a reference without a large number of repeated tests, the fault tolerance of instrument analysis is improved, for example, if a waveform mutation exists in the waveform of item A of a substance, the waveform mutation is recorded in the instrument model, and the waveform mutation can be determined as the actual component of A. Further analysis through other devices or higher-priced similar devices can be avoided, and the cost is reduced. Of course, the above is only a typical example of the present application, and in addition to this, the present application can have other various specific embodiments, and any technical solution formed by equivalent replacement or equivalent transformation falls within the scope of the present application.

Claims

1. A chemical instrument follow-up analysis system, characterized in that: It includes instrument model construction subsystem, experimental data analysis subsystem and instrument model correction subsystem; The instrument model construction subsystem is used to generate an instrument model, which includes a plurality of modular instrument information. The modular instrument information includes instrument content data, instrument feature data, and analysis type data. The instrument content data reflects the type of chemical instrument, the instrument feature data reflects the analysis features of the chemical instrument, and the analysis type data reflects the analysis scenario applicable to the chemical instrument. The instrument model construction subsystem includes a sample input module, a feature network construction module, and an analysis network construction module. The sample input module inputs different chemical instrument information according to different chemical instrument types and processes the input chemical instrument information in a fixed format to generate modular information. The feature network construction module is configured with a preset feature extraction strategy and a plurality of feature association conditions. The feature extraction strategy is used to extract instrument features of the instrument feature data, and the feature association conditions are used to determine the association relationship between the instrument features in different modular instrument information to generate a feature association index. The analysis network construction module is configured with a preset type extraction strategy and a plurality of type association conditions. The type extraction strategy is used to extract the analysis type of the analysis type data, and the analysis type index is generated based on modular instrument information with the same analysis type. The experimental data analysis subsystem includes an information acquisition module, an information indexing module, and an information analysis module. The information acquisition module is used to obtain experimental project information, and the experimental project information includes experimental content data and experimental requirement data. The information indexing module extracts corresponding instrument characteristics based on the experimental content data in the experimental project information, and determines the corresponding analysis type based on the experimental requirement data. The corresponding modular instrument information is determined from the instrument model based on the instrument characteristics and analysis type. The information analysis module screens the modular instrument information based on the experimental requirement data and outputs instrument report information based on the obtained modular instrument information. The instrument model correction subsystem includes a correction trigger module and an information correction module. The correction trigger module is configured with a correction trigger condition. When the experimental project information meets the correction trigger condition, the corresponding experimental content data is extracted; The information correction module generates new instrument feature data according to the experimental content data and updates the instrument model. Meanwhile, the feature network construction module updates the corresponding feature association index according to the new instrument feature data.

2. A chemical instrument follow-up analysis system according to claim 1, characterized in that: The system further includes an abnormality analysis subsystem, wherein the chemical instrument information includes standard result information and abnormal result information, wherein the standard result information reflects the analysis result of the instrument under normal conditions, and the abnormal result information reflects the analysis result of the instrument under abnormal conditions, wherein the sample input module compares the standard result information with the abnormal result information to generate abnormal feature items, and generates the instrument feature data according to the abnormal feature items, wherein the abnormal feature items include abnormal feature content and abnormal feature type; The abnormality analysis subsystem includes an abnormality identification module and an abnormality analysis module; the abnormality identification module determines whether the same abnormal feature content exists in the modular information based on the detection results in the experimental project information. If the same abnormal feature content exists, the abnormal feature type is sent to the abnormality analysis module. The abnormality analysis module retrieves the corresponding modular instrument information based on the abnormal feature content and outputs abnormality report information based on the obtained modular instrument information and abnormal feature type.

3. A chemical instrument follow-up analysis system according to claim 2, characterized in that: The sample input module also includes a behavior analysis unit, which is connected to the analysis and collection end. The analysis and collection end is used to collect experimental behavior information, which reflects the user's behavior during the experiment. The behavior analysis unit parses the experimental behavior information to generate the abnormal feature type.

4. A chemical instrument follow-up analysis system according to claim 3, characterized in that: The abnormality analysis subsystem further includes a data compensation module, which generates a compensation strategy according to the abnormality feature type and corrects the detection result using the compensation strategy.

5. A chemical instrument follow-up analysis system as claimed in claim 3, characterized in that: The analysis and acquisition end is configured as an image collector, which is arranged facing the experimental space of the chemical instrument. The behavior analysis unit generates the abnormal feature type based on the acquired image information.

6. A chemical instrument follow-up analysis system according to claim 5, characterized in that: The behavior analysis unit is configured with a behavior analysis strategy, which parses standard behavior features to form an image verification sequence. The image verification sequence includes several image verification features, and compares the image information in turn to see whether the corresponding image verification features exist. If not, the image verification feature is marked and the abnormal feature type is generated based on the marked image verification feature.

7. A chemical instrument follow-up analysis system according to claim 6, characterized in that: The standard behavior characteristics are generated according to the experimental project information.

8. A chemical instrument follow-up analysis system according to claim 6, characterized in that: The standard behavior characteristics are generated based on instrument characteristic data.

9. A chemical instrument follow-up analysis system as claimed in claim 3, characterized in that: The analysis and collection end is a state monitoring unit, which is connected to a corresponding chemical instrument and monitors the working state of the chemical instrument to generate the abnormal feature type.

Images