Wheat flour vomitoxin residue analysis system based on data analysis

Abstract

The invention relates to the field of food safety detection, and discloses a wheat flour vomitoxin residue analysis system based on data analysis. Comprising a sample acquisition and preprocessing module, a vomitoxin detection module, a data acquisition and storage module, a data analysis and modeling module, an intelligent early warning and decision support module, a visual display module and a compliance and report generation module. The vomitoxin detection module is used for receiving a sample from the sample acquisition and preprocessing module; the data acquisition and storage module is connected with the sample acquisition and preprocessing module, the vomitoxin detection module, the data analysis and modeling module and the intelligent early warning and decision support module. By arranging the automatic multi-point sampling device, representative sampling can be automatically carried out according to different links of production batches and supply chains. Automatic sampling not only improves the frequency and accuracy of sample collection, but also reduces human intervention, significantly improves the sampling efficiency, and is especially suitable for large-scale production scenes.

Description

Technical Field

[0001] The invention relates to the field of food safety detection, and in particular to a wheat flour vomitoxin residue analysis system based on data analysis. Background Art

[0002] Deoxynivalenol (DON) is a type of secondary metabolite produced by fungi and is commonly found in wheat and other cereals infected with fungi. Due to its threat to human and animal health, the global food and feed industry pays special attention to the detection and monitoring of DON in cereals. Existing methods for analyzing DON residues in wheat flour mainly rely on laboratory testing techniques, which usually involve manual sampling, laboratory chemical analysis, and manual data processing. However, these existing technologies have some obvious drawbacks in practical applications, which limits their adaptability and efficiency in large-scale production environments.

[0003] Existing systems usually rely on manual sampling, which cannot guarantee the representativeness of samples, especially in large-scale production and multi-batch supply chains, where the randomness and frequency of sample collection are limited. In addition, the manual sampling process is cumbersome and susceptible to human factors, resulting in low sample collection efficiency, especially under high production demand, manual sampling is difficult to adapt quickly. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a wheat flour vomitoxin residue analysis system based on data analysis, which solves the problems that the existing system usually relies on manual sampling and cannot guarantee the representativeness of the samples, especially in large-scale production and multi-batch supply chains, the randomness of samples and the limited collection frequency.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a wheat flour vomitoxin residue analysis system based on data analysis, including a sample collection and pretreatment module, a vomitoxin detection module, a data collection and storage module, a data analysis and modeling module, an intelligent early warning and decision support module, a visualization display module, and a compliance and report generation module; The sample collection and pretreatment module is directly connected to the vomitoxin detection module. At the same time, the sample collection and pretreatment module is connected to the data collection and storage module through a data interface to transmit the sample number, sampling time and location; The DON detection module receives samples from the sample collection and preprocessing module and transmits the test results to the data collection and storage module through an interface. At the same time, it communicates with the data analysis and modeling module to provide real-time test data for analysis and prediction; The data acquisition and storage module is connected to the sample collection and preprocessing module, the vomitoxin detection module, the data analysis and modeling module, and the intelligent early warning and decision support module, and serves as a data transmission hub between all modules; The data analysis and modeling module is connected to the data acquisition and storage module to receive the detection data for analysis. At the same time, the data analysis and modeling module is connected to the intelligent early warning and decision support module to provide the analysis results and prediction model to the intelligent early warning and decision support module for triggering early warnings and generating management decisions; The intelligent early warning and decision support module is closely connected with the data analysis and modeling module, relying on the results of the data analysis and modeling module to trigger early warnings and provide decision support based on the predictions generated by the model. At the same time, the early warning and decision information are delivered to users through the visual display module; The visualization display module obtains data from the data acquisition and storage module, the data analysis and modeling module, and the intelligent early warning and decision support module to generate visualization content, and is connected to the compliance and report generation module to generate and output reports that comply with food safety standards; The compliance and report generation module is connected with the data collection and storage module, the data analysis and modeling module, and the visualization display module to obtain relevant data to generate reports. The reports generated by the compliance and report generation module can be exported and archived or reviewed by users.

[0006] Preferably, the sample collection and pretreatment module includes an automated sample collection device, which has a multi-point sampling function and transmits the collected samples to a physical crushing device through a conveying device. After the physical crushing device evenly crushes the sample to a set particle size, the crushed sample is mixed with a chemical solvent in a preset proportion to extract vomitoxin, and the chemical solvent includes acetic acid or methanol.

[0007] Preferably, the vomitoxin detection module includes a high performance liquid chromatography-mass spectrometry device and an enzyme-linked immunosorbent assay device, wherein the high performance liquid chromatography-mass spectrometry device is used to detect the content of vomitoxin by separation and mass spectrometry analysis, and the enzyme-linked immunosorbent assay device is used to screen vomitoxin based on antigen-antibody reaction, and the vomitoxin detection module is connected to the sample pretreatment module via an interface to realize automated sampling and detection operations.

[0008] Preferably, the data acquisition and storage module includes a real-time data acquisition unit and a distributed database storage unit. The real-time data acquisition unit automatically collects detection data from the detection equipment through an API interface, including sample number, detection time and toxin content. The distributed database storage unit classifies and stores the collected data according to sample number and detection date, and supports multi-node data backup and fast reading.

[0009] Preferably, the data analysis and modeling module includes a data preprocessing unit, a model training unit and a prediction model generation unit, the data preprocessing unit normalizes, denoises and detects outliers on the detection data, the model training unit is based on a machine learning algorithm, including a random forest, a support vector machine and a deep learning model, and the prediction model generation unit generates future DON residue prediction results based on the trained model.

[0010] Preferably, the intelligent early warning and decision support module includes a data processing unit, an early warning generation unit and a decision support unit. The data processing unit generates a threshold value based on the analysis results of the data analysis and modeling module. When the detection result or trend exceeds the threshold value, the early warning generation unit triggers an early warning signal. The decision support unit generates corresponding decision parameters based on historical data and prediction results for production adjustment and management reference.

[0011] Preferably, the visualization display module includes a dynamic data display unit and a report generation unit, wherein the dynamic data display unit displays the test results and DON residue trend information through a dashboard and charts, and the report generation unit generates a customized report containing the test data and analysis results based on user selection, and supports exporting to PDF or Excel files.

[0012] Preferably, the compliance and report generation module includes a compliance update unit and an audit record unit. The compliance update unit dynamically tracks the updates of national and international food safety standards and adjusts the detection and report generation rules accordingly. The audit record unit records all steps in the detection process, including timestamps, operator information and equipment parameters.

[0013] Preferably, a working method of a wheat flour vomitoxin residue analysis system based on data analysis comprises the following steps: S1: Sample collection and preprocessing stage: The automated sample collection device obtains wheat flour samples from different production batches or supply chain links according to a preset multi-point sampling plan; The collected samples are transported to the physical crushing equipment through a conveyor device. After being crushed to a set particle size, they are automatically mixed with a chemical solvent in a predetermined proportion to extract the vomitoxin in the sample; The sample number, sampling time and location information are transmitted to the data acquisition and storage module through the data interface for recording; S2: DON detection stage: The pre-treated samples are automatically transferred to the DON detection module. The module uses a high-performance liquid chromatography-mass spectrometry device to accurately detect the amount of vomitoxin and an enzyme-linked immunosorbent assay device to perform large-scale screening; During the testing process, the testing equipment automatically transmits the test results to the data acquisition and storage module, and passes the real-time test data to the data analysis and modeling module; S3: Data collection and storage stage: The data collection and storage module collects the test data of samples in real time through the API interface, including sample number, test time and toxin content; The collected data is classified and stored according to the sample number, sampling time and test date, and the data is backed up through a distributed database; S4: Data analysis and modeling stage: The data analysis and modeling module preprocesses the detection data received from the data acquisition and storage module, including data normalization, denoising, and outlier detection; The preprocessed data were used to train a machine learning model to establish a prediction model for DON residues; The data analysis and modeling module generates future prediction results of DON residues based on historical data and the latest test data, and passes the analysis results to the intelligent early warning and decision support module; S5: Intelligent early warning and decision support stage: The intelligent early warning and decision support module sets the early warning threshold according to the analysis results and prediction models generated by the data analysis and modeling module; When the detection data or prediction results exceed the set threshold, the warning generation unit triggers the warning signal and generates decision parameters for production adjustment or supply chain management through the data processing unit; Early warning and decision-making information is delivered to users through the visual display module; S6: Visualization display and report generation stage: The visualization display module obtains data from the data acquisition and storage module, the data analysis and modeling module, and the intelligent early warning and decision support module, and displays the test results, DON residue trends, and early warning information to the user in the form of charts or dashboards; Users can generate customized reports through the visualization module and choose to export them as PDF or Excel files; S7: Compliance and Report Generation Phase: The compliance and reporting module dynamically tracks updates to national and international food safety standards; The Compliance and Report Generation module automatically generates detailed test reports containing sample information, test results and analytical data, while recording audit information for each test step, including timestamps, operator information and equipment parameters.

[0014] The present invention provides a wheat flour vomitoxin residue analysis system based on data analysis, which has the following beneficial effects: 1. The present invention provides a wheat flour vomitoxin residue analysis system based on data analysis, which is equipped with an automated multi-point sampling device and can automatically perform representative sampling according to production batches and different links of the supply chain. Automated sampling not only improves the frequency and accuracy of sample collection, but also reduces human intervention, significantly improves sampling efficiency, and is particularly suitable for large-scale production scenarios.

[0015] 2. This system integrates high performance liquid chromatography-mass spectrometry (HPLC-MS) and enzyme-linked immunosorbent assay (ELISA), which can directly perform real-time detection after sampling, and automatically transmit the test results to the data storage module through the data interface. Compared with traditional laboratory testing methods, this invention shortens the detection cycle, allowing the production line to make rapid adjustments based on real-time data, avoiding the problem of delayed detection.

[0016] 3. The system of the present invention performs unified test data management through a distributed database. All sample numbers, sampling times and test results can be automatically stored and backed up to ensure data integrity and traceability. The data analysis module can automatically process test data, perform data normalization, denoising and trend analysis, so that managers can quickly grasp the food safety dynamics in the supply chain.

[0017] 4. The system of the present invention uses machine learning algorithms and historical data to predict and analyze future DON residues, and can trigger an early warning before the DON content of high-risk batches or suppliers reaches a dangerous level. The early warning module sends early warning signals in real time according to the set threshold, providing managers with the opportunity to intervene in advance and effectively reducing the risk of food safety accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a diagram of the architecture of a wheat flour vomitoxin residue analysis system based on data analysis according to the present invention; Figure 2 The present invention is a flow chart of the working method of a wheat flour vomitoxin residue analysis system based on data analysis. DETAILED DESCRIPTION

[0019] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Please see attached Figure 1 - Attachment Figure 2, an embodiment of the present invention provides a wheat flour vomitoxin residue analysis system based on data analysis, including a sample collection and pretreatment module, a vomitoxin detection module, a data collection and storage module, a data analysis and modeling module, an intelligent early warning and decision support module, a visualization display module, and a compliance and report generation module; The sample collection and pretreatment module is directly connected to the vomitoxin detection module, and the sample is automatically transmitted to the detection device through the transmission device. At the same time, the sample collection and pretreatment module is connected to the data collection and storage module through the data interface, and automatically transmits metadata related to each sample, including key information such as sample number, sampling time, sampling location, etc., which are used for subsequent data processing and analysis.

[0021] The DON detection module receives samples from the sample collection and pretreatment module and analyzes the DON content according to the set detection process. The test results are transmitted to the data acquisition and storage module through a standard interface for subsequent data analysis and storage. This module is also connected to the data analysis and modeling module to ensure that real-time test data is available during the analysis process for machine learning models to use for training and prediction.

[0022] The data acquisition and storage module is connected to the sample collection and preprocessing module, the vomitoxin detection module, the data analysis and modeling module, and the intelligent early warning and decision support module, and serves as the data transmission hub between all modules. It receives and stores data transmitted by each module in real time, adopts a distributed database architecture to ensure the security, reliability and fast access of data storage, and provides multi-node data backup support.

[0023] The data analysis and modeling module is connected to the data acquisition and storage module to receive the test data and perform standardized processing, including denoising and data cleaning. The preprocessed data is sent to the model training unit and trained using machine learning algorithms (such as random forest, support vector machine and deep learning). The generated prediction model is used to predict the future trend of DON residues, and the prediction results and analysis results are transmitted to the intelligent early warning and decision support module through the interface.

[0024] The intelligent early warning and decision support module is closely connected with the data analysis and modeling module, receiving and processing the prediction results and analysis data in real time. This module relies on the results of the data analysis and modeling module to set the early warning threshold. When the detection result or prediction result exceeds the set threshold, the early warning generation unit automatically triggers the early warning signal to inform the relevant person in charge. The decision support unit uses historical data and current analysis to generate operational suggestions to provide a basis for production adjustments and management decisions. The early warning information is transmitted to the user through the visual display module.

[0025] The visualization module obtains real-time data from the data acquisition and storage module, the data analysis and modeling module, and the intelligent early warning and decision support module to generate dashboards, charts and other visualization content to display the test results and trend information of DON residues. Users can interact with the data through the interface to generate and customize reports. This module is also connected to the compliance and report generation module to generate reports that comply with national and international food safety standards.

[0026] The compliance and report generation module is connected with the data acquisition and storage module, data analysis and modeling module, and visualization module to obtain relevant data and generate reports. This module includes the compliance update unit and the audit record unit. The compliance update unit continuously monitors changes in national and international food safety standards and automatically adjusts the report generation rules. The audit record unit records detailed information of all relevant steps in the testing process, including timestamps, operator information, equipment parameters, etc., to ensure data transparency and traceability.

[0027] The sample collection and pretreatment module includes an automated sample collection device with a multi-point sampling function, and transmits the collected samples to a physical crushing device through a conveyor. The samples are evenly crushed to a set particle size standard in the crushing device, and the crushed samples are automatically mixed with a preset proportion of chemical solvents (such as acetic acid or methanol) to extract vomitoxin. The design of this module makes the sample pretreatment process fully automated, reducing human errors.

[0028] The DON detection module includes a HPLC-MS device and an ELISA device. The HPLC-MS device is responsible for accurate quantitative detection of DON in samples through liquid phase separation and mass spectrometry analysis technology, and is suitable for detection of toxin residues at lower concentrations. The ELISA device achieves batch screening through antigen-antibody reaction, and is suitable for rapid initial screening. The detection process is automated to ensure efficient sample processing and transmission of test results.

[0029] The data collection and storage module includes a real-time data collection unit and a distributed database storage unit. The real-time data collection unit automatically collects relevant data from the vomitoxin detection module and the sample collection and pretreatment module through the API interface, including sample number, detection time, vomitoxin content, etc. The distributed database storage unit classifies and stores the collected data, indexes it by metadata such as sample number and detection date, supports fast reading and query, and provides multi-node data redundant backup to ensure data security.

[0030] The data analysis and modeling module includes a data preprocessing unit, a model training unit, and a prediction model generation unit. The data preprocessing unit first normalizes the test data to remove noise and outliers to ensure data stability. The model training unit models historical data based on machine learning algorithms and uses random forests, support vector machines, and deep learning models to predict and analyze the DON residues in samples. The prediction model generation unit applies the optimized model to new data to predict future residue trends.

[0031] The intelligent early warning and decision support module includes a data processing unit, an early warning generation unit, and a decision support unit. The data processing unit is responsible for generating dynamic thresholds based on the analysis results of the data analysis and modeling module. When the residual data of vomitoxin or the predicted trend exceeds the set threshold, the early warning generation unit triggers the early warning signal in the system and sends a notification to the relevant person in charge. The decision support unit generates decision parameters for adjusting production or supply chain management based on historical data and model prediction results, and provides guiding suggestions.

[0032] The visualization module includes a dynamic data display unit and a report generation unit. The dynamic data display unit displays the test results and trend information of vomitoxin through multi-dimensional dashboards and interactive charts. Users can filter data according to their needs and generate corresponding analysis result charts. The report generation unit generates customized reports containing detailed test data and analysis conclusions based on user input or automatic configuration, and supports export in PDF and Excel formats for easy archiving and sharing.

[0033] The compliance and report generation module includes the compliance update unit and the audit record unit. The compliance update unit monitors the updates of national and international food safety regulations in real time, automatically adjusts the generation rules of the test report, and ensures that the report content always complies with the latest laws and standards. The audit record unit is responsible for recording all key steps in the test process, including operator information, timestamps, equipment operating status, etc., and generates detailed audit logs for future review and traceability.

[0034] A working method of a wheat flour vomitoxin residue analysis system based on data analysis comprises the following steps: S1: Sample collection and preprocessing stage Sample collection: The automated sample collection device obtains wheat flour samples from multiple production batches and different links of the supply chain according to the preset sampling plan. The sampling device integrates multi-point sampling function, which can automatically perform sampling tasks according to the preset time, location and batch to ensure that the sample is representative. The sampling plan is set according to the system's built-in algorithm, taking into account variables such as production batch size, storage time, processing technology, etc., to ensure coverage of key nodes in the entire supply chain.

[0035] Sample transfer and pretreatment: The collected samples are transported to the pretreatment unit through an automated conveyor, and first enter the physical crushing device, where the samples are evenly crushed to a set particle size standard (such as less than 0.5 mm). The purpose of physical crushing is to ensure that the vomitoxin can be completely released in the subsequent extraction step to improve the detection accuracy. After the crushing is completed, the sample automatically enters the chemical extraction unit, where the sample is fully mixed with a preset proportion of chemical solvents (such as acetic acid or methanol). The choice and proportion of the solvent are pre-optimized according to the type and concentration characteristics of vomitoxin to ensure extraction efficiency and accuracy.

[0036] Data collection and transmission: After the sample is collected, the sample number, sampling time, location and other information are automatically transmitted to the data collection and storage module through the integrated sensor and data interface for recording. This information is encrypted to ensure data integrity and security during transmission and storage.

[0037] S2: DON detection stage Sample test preparation: The pre-treated sample is directly transferred to the DON test module via an automated sample transfer system. Here, the system automatically selects the test method (HPLC-MS or ELISA) based on the sample type and target toxin concentration. The test conditions of the sample (such as solvent gradient, flow rate, etc.) are automatically adjusted to ensure optimal test performance.

[0038] High performance liquid chromatography-mass spectrometry (HPLC-MS) detection: The sample is separated by the HPLC system and then enters the mass spectrometer for precise analysis of the vomitoxin components. The advantage of mass spectrometry is that it can quantitatively detect extremely low concentrations of vomitoxin (ppb level) and has high-resolution detection capabilities for many different types of vomitoxin. During the detection process, the system automatically records the operating information of each detection step, including sample flow rate, separation column temperature, mass spectrometry parameters, etc., and transmits these parameters in real time to the data acquisition and storage module for recording.

[0039] Enzyme-linked immunosorbent assay (ELISA) test: For the initial screening of large batches of samples, the system uses the ELISA method. This method is based on the antigen-antibody reaction and can quickly screen for the presence of vomitoxin. Through the automated microplate processing system, the system can process hundreds of samples at a time, significantly improving the speed and efficiency of testing. The ELISA test results are transmitted to the data acquisition and storage module through an automatic interface for subsequent analysis.

[0040] Detection data transmission: During the detection process, all test results, including DON content, detection time, equipment parameters and other information, are transmitted to the data acquisition and storage module in real time, and simultaneously passed to the data analysis and modeling module for further data analysis and modeling processing.

[0041] S3: Data collection and storage stage Real-time data collection: The data acquisition and storage module is seamlessly connected to the testing equipment (HPLC-MS or ELISA) through the API interface to receive all test data in real time. This module not only collects test results, but also collects the operating status of the equipment, operator information, environmental parameters, etc. to ensure the comprehensiveness and traceability of the data.

[0042] Data classification and storage: All data is automatically classified and stored according to multiple dimensions (sample number, test time, sampling location, etc.). During the storage process, the system adopts a distributed database architecture, and the data is distributed to multiple nodes for redundant backup to ensure the integrity and availability of the data under any hardware failure. At the same time, the system supports high-speed reading and query functions to ensure that data can be quickly extracted during analysis or auditing.

[0043] Data encryption and security: The system uses multi-layer encryption technology to protect all detection data, including data transmission encryption, storage encryption and access control, to ensure that unauthorized users cannot access or modify the data.

[0044] S4: Data analysis and modeling stage Data preprocessing: The data analysis and modeling module first preprocesses the test data received from the data acquisition and storage module, including data normalization, denoising, outlier detection and data cleaning. Normalization ensures that the data is consistent under different test conditions, while denoising and outlier processing eliminate noise and abnormal data in the test process to improve the accuracy of the analysis.

[0045] Machine Learning Model Training: The preprocessed data is input into the model training unit, and the model is trained using a variety of machine learning algorithms (such as random forest, support vector machine, and deep learning) based on historical detection data. The system automatically adjusts the training parameters to ensure that the generated model has high accuracy and high generalization ability. During the model training process, the system will cross-validate and evaluate the performance of the model to ensure its stability on different data sets.

[0046] Prediction model generation and analysis result output: The trained model is used to predict the trend of DON residue in the future. The data analysis and modeling module combines historical data with the latest test data to generate detailed prediction results, including the concentration trend of residual toxins, potential high-risk samples, etc. The prediction results and analysis results are passed to the intelligent early warning and decision support module through the data interface.

[0047] S5: Intelligent early warning and decision support stage Warning threshold setting: The intelligent early warning and decision support module automatically sets the early warning threshold based on the analysis results generated by the data analysis and modeling module. The threshold is set based on national food safety standards and statistical analysis results of historical data to ensure that the system can issue early warnings in a timely manner when toxin residues exceed the standard or trends are abnormal.

[0048] Warning triggering and decision making: When the test data or prediction results exceed the preset threshold, the warning generation unit automatically triggers the warning signal and notifies the relevant responsible personnel through the system. At the same time, the decision support unit generates optimization suggestions for production adjustment or supply chain management based on the warning information, including suspension of production, adjustment of raw material supply, re-testing and other operating parameters.

[0049] Decision delivery and execution: The generated warning and decision parameters are transmitted to users through the visual display module. Users can view the warning details through the interface and adjust the production process or supply chain based on the decision suggestions.

[0050] S6: Visualization and report generation stage Data display: The visualization module obtains real-time data from the data acquisition and storage module, the data analysis and modeling module, and the intelligent early warning and decision support module to generate multi-dimensional visualization charts and dashboards. Users can view the test results, residual trends and early warning information of vomitoxin in real time. The system supports dynamic data interaction and multi-dimensional analysis, such as filtering by time period, sample source or batch.

[0051] Report Generation: The visualization module allows users to customize the report format. Users can select different data dimensions and analysis results according to their needs to generate reports that meet specific needs. The report generation module supports multiple export formats, including PDF and Excel, which facilitates data archiving, sharing or submission to regulatory authorities.

[0052] S7: Compliance and reporting phase Compliance updates and testing adjustments: The Compliance and Report Generation module continuously monitors updates to national and international food safety standards and automatically adjusts the testing process and report generation rules based on the latest standards. The system automatically corrects the testing thresholds and report content based on updates to laws and regulations to ensure that the generated reports meet the latest compliance requirements.

[0053] Detailed report generation and audit trails: The system automatically generates detailed reports containing sample information, test results, data analysis results and trend forecasts, including audit records of each test step (such as timestamp, operator information, equipment status, etc.). All test data and operation records are encrypted to ensure their integrity and traceability. The generated reports can be directly exported and archived, reviewed or submitted to third-party regulatory agencies.

[0054] The following is an introduction in conjunction with specific embodiments: Example 1: DON Monitoring and Analysis for Large-Scale Wheat Flour Production background: A large flour production company processes wheat flour from different suppliers in multiple production workshops. In order to ensure the food safety of the final product, the company needs to test each batch of wheat flour for DON residues and make adjustments and optimizations based on the test results.

[0055] process: S1: Sample collection and preprocessing stage Sampling operation: Each production line of the enterprise is equipped with an automated sample collection device, which automatically collects samples from each batch of wheat flour provided by each supplier according to the sampling plan set by the system. In this embodiment, the system randomly selects 5 time points from different batches on each production line for sampling every day.

[0056] Sample pretreatment: The collected samples are automatically transferred to the physical crushing equipment through the conveyor, crushed to a uniform particle size of 0.2 mm, and then mixed with a preset proportion of acetic acid solvent for vomitoxin extraction. The number, sampling time and production batch information of each sample are automatically recorded in the data acquisition and storage module.

[0057] S2: DON detection stage Detection operation: The pre-treated samples are transferred to the DON detection module. The system uses a high performance liquid chromatography-mass spectrometry (HPLC-MS) device to accurately detect DON in the sample. It takes about 15 minutes to measure the DON content of each sample. For large batches of samples, the system uses enzyme-linked immunosorbent assay (ELISA) for rapid screening to quickly identify batches that may have higher toxin levels.

[0058] Data transmission: The test results of each sample are transmitted to the data acquisition and storage module in real time through the interface, and are also passed to the data analysis and modeling module for subsequent analysis.

[0059] S3: Data collection and storage stage Data storage operations: The data collection and storage module receives test data in real time through the API interface, and all sample numbers, sampling time, vomitoxin content and other information are classified and stored. The data is indexed according to production batch and sampling time and stored in a distributed database. The system automatically generates redundant backups to ensure data security and reliability.

[0060] S4: Data analysis and modeling stage Data analysis operations: The system normalizes the data of each batch of DON testing, removes outliers and noise, and then models the data using a machine learning model based on random forests and support vector machines. The model combines historical data to predict the trend of DON residues in future production batches. In this example, the system predicts that a batch of wheat flour provided by a supplier may have a higher risk of DON residues in the future.

[0061] S5: Intelligent early warning and decision support stage Early warning and decision-making operations: The intelligent early warning and decision support module sets the early warning threshold for the concentration of vomitoxin based on the analysis results. For some batches of wheat flour from certain suppliers, the system predicts that the residue of vomitoxin in future production may exceed the safety standard. Therefore, the system triggers an early warning signal and recommends additional testing of the batch of wheat flour. The decision support unit generates a recommendation for the production manager to immediately suspend the supply of raw materials from this supplier until further testing is completed.

[0062] S6: Visualization and report generation stage Report and display operations: The visualization module provides the production management team with real-time data of all production batches, including the test results and early warning information of vomitoxin for each batch. Through the charts generated by the system, users can clearly see the quality trends of raw materials from different suppliers. The system allows users to generate customized PDF reports based on the test results, which list the sample number, test data and analysis results in detail and export them as electronic files for review by regulatory authorities.

[0063] S7: Compliance and reporting phase Compliance management operations: The compliance and report generation module automatically generates test reports that comply with national food safety standards and dynamically tracks the latest regulatory updates. In this example, the system automatically adjusts the test process and report generation rules based on the latest international food standards. Each report records audit information of all operation steps, including test time, operator ID, equipment operating status, etc., for subsequent review.

[0064] Example 2: Intelligent monitoring and early warning system for small flour mills background: A small flour mill processes wheat flour from multiple suppliers every day, but its resources are limited and it is unable to continuously monitor the DON residue in each batch. By introducing a residue analysis system based on data analysis, the plant hopes to improve the efficiency and accuracy of food safety management.

[0065] process: S1: Sample collection and preprocessing stage Sampling operation: The system is equipped with a simplified automated sample collection device for the production line of a small factory, collecting 3 samples from each key batch of incoming materials every day. Due to the small number of batches processed, the sampling plan is relatively simplified, but the system still ensures the comprehensiveness and accuracy of the sampling. The sample number, sampling time, and supplier information are automatically stored.

[0066] S2: DON detection stage Detection operation: The plant chose to use ELISA for rapid screening to detect the presence of vomitoxin. When a batch is screened as high risk, the system transfers the batch sample to HPLC-MS for detailed toxin content analysis. The test results are transmitted to the data storage module in real time through an automatic interface and passed to the data analysis module.

[0067] S3: Data collection and storage stage Data storage operations: The system receives the inspection data of each batch through the API interface. The data is stored in a distributed database by supplier, time and batch number, and the factory can access and back up the data at any time.

[0068] S4: Data analysis and modeling stage Data analysis operations: The system modeled historical data, focusing specifically on the quality of raw materials from a specific supplier. The analysis showed that a supplier had a trend of higher DON residues in its products over the past three months.

[0069] S5: Intelligent early warning and decision support stage Early warning and decision-making operations: Based on the model prediction, the system generated a risk warning of residual vomitoxin from the supplier, and recommended suspending the use of the supplier's raw materials and arranging additional testing. On the management interface, the factory's operations manager can view the warning details in real time and make decisions based on the system's recommendations.

[0070] S6: Visualization and report generation stage Demonstration operation: The system displays the test data and analysis results of each batch through a visual dashboard. Operations managers can filter data by supplier, date or batch and generate PDF or Excel reports for internal review and submission to customers.

[0071] S7: Compliance and reporting phase Compliance Report Operations: The system generates detailed reports based on the food safety regulations required by the factory. Each report contains an audit record of all test steps, including equipment parameters, test time and operator information, ensuring that the report meets food safety regulatory requirements.

[0072] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A wheat flour vomitoxin residue analysis system based on data analysis, characterized in that: It includes sample collection and pretreatment module, vomitoxin detection module, data collection and storage module, data analysis and modeling module, intelligent early warning and decision support module, visualization display module, and compliance and report generation module; The sample collection and pretreatment module is directly connected to the vomitoxin detection module. At the same time, the sample collection and pretreatment module is connected to the data collection and storage module through a data interface to transmit the sample number, sampling time and location; The DON detection module receives samples from the sample collection and preprocessing module and transmits the test results to the data collection and storage module through an interface. At the same time, it communicates with the data analysis and modeling module to provide real-time test data for analysis and prediction; The data acquisition and storage module is connected to the sample collection and preprocessing module, the vomitoxin detection module, the data analysis and modeling module, and the intelligent early warning and decision support module, and serves as a data transmission hub between all modules; The data analysis and modeling module is connected to the data acquisition and storage module to receive the detection data for analysis. At the same time, the data analysis and modeling module is connected to the intelligent early warning and decision support module to provide the analysis results and prediction model to the intelligent early warning and decision support module for triggering early warnings and generating management decisions; The intelligent early warning and decision support module is closely connected with the data analysis and modeling module, relying on the results of the data analysis and modeling module to trigger early warnings and provide decision support based on the predictions generated by the model. At the same time, the early warning and decision information are delivered to users through the visual display module; The visualization display module obtains data from the data acquisition and storage module, the data analysis and modeling module, and the intelligent early warning and decision support module to generate visualization content, and is connected to the compliance and report generation module to generate and output reports that comply with food safety standards; The compliance and report generation module is connected with the data collection and storage module, the data analysis and modeling module, and the visualization display module to obtain relevant data to generate reports. The reports generated by the compliance and report generation module can be exported and archived or reviewed by users.

2. The wheat flour vomitoxin residue analysis system based on data analysis according to claim 1, characterized in that: The sample collection and pretreatment module includes an automated sample collection device, which has a multi-point sampling function and transmits the collected samples to a physical crushing device through a transmission device. After the physical crushing device evenly crushes the sample to a set particle size, the crushed sample is mixed with a preset proportion of a chemical solvent to extract vomitoxin, and the chemical solvent includes acetic acid or methanol.

3. The wheat flour vomitoxin residue analysis system based on data analysis according to claim 1, characterized in that: The vomitoxin detection module includes a high performance liquid chromatography-mass spectrometry device and an enzyme-linked immunosorbent assay device, wherein the high performance liquid chromatography-mass spectrometry device is used to detect the content of vomitoxin by separation and mass spectrometry analysis, and the enzyme-linked immunosorbent assay device is used to screen vomitoxin based on antigen-antibody reaction. The vomitoxin detection module is connected to the sample pretreatment module through an interface to realize automated sampling and detection operations.

4. The wheat flour vomitoxin residue analysis system based on data analysis according to claim 1, characterized in that: The data acquisition and storage module includes a real-time data acquisition unit and a distributed database storage unit. The real-time data acquisition unit automatically collects detection data from the detection equipment through an API interface, including sample number, detection time and toxin content. The distributed database storage unit classifies and stores the collected data according to sample number and detection date, and supports multi-node data backup and fast reading.

5. The wheat flour vomitoxin residue analysis system based on data analysis according to claim 1, characterized in that: The data analysis and modeling module includes a data preprocessing unit, a model training unit and a prediction model generation unit. The data preprocessing unit normalizes, denoises and detects outliers on the detection data. The model training unit is based on a machine learning algorithm, including a random forest, a support vector machine and a deep learning model. The prediction model generation unit generates future prediction results of vomitoxin residues based on the trained model.

6. The wheat flour vomitoxin residue analysis system based on data analysis according to claim 1, characterized in that: The intelligent early warning and decision support module includes a data processing unit, an early warning generation unit and a decision support unit. The data processing unit generates a threshold value according to the analysis results of the data analysis and modeling module. When the detection result or trend exceeds the threshold value, the early warning generation unit triggers an early warning signal. The decision support unit generates corresponding decision parameters based on historical data and prediction results for production adjustment and management reference.

7. The wheat flour vomitoxin residue analysis system based on data analysis according to claim 1, characterized in that: The visualization display module includes a dynamic data display unit and a report generation unit. The dynamic data display unit displays the test results and DON residue trend information through dashboards and charts. The report generation unit generates a customized report containing test data and analysis results based on user selections, and supports exporting to PDF or Excel files.

8. The wheat flour vomitoxin residue analysis system based on data analysis according to claim 1, characterized in that: The compliance and report generation module includes a compliance update unit and an audit record unit. The compliance update unit dynamically tracks the updates of national and international food safety standards and adjusts the detection and report generation rules accordingly. The audit record unit records all steps in the detection process, including timestamps, operator information and equipment parameters.

9. A working method of a wheat flour vomitoxin residue analysis system based on data analysis, characterized in that: The wheat flour vomitoxin residue analysis system based on data analysis for any one of claims 1 to 8 comprises the following steps: S1: Sample collection and preprocessing stage: The automated sample collection device obtains wheat flour samples from different production batches or supply chain links according to a preset multi-point sampling plan; The collected samples are transported to the physical crushing equipment through a conveyor device. After being crushed to a set particle size, they are automatically mixed with a chemical solvent in a predetermined proportion to extract the vomitoxin in the sample; The sample number, sampling time and location information are transmitted to the data acquisition and storage module through the data interface for recording; S2: DON detection stage: The pre-treated samples are automatically transferred to the DON detection module. The module uses a high-performance liquid chromatography-mass spectrometry device to accurately detect the amount of vomitoxin and an enzyme-linked immunosorbent assay device to perform large-scale screening; During the testing process, the testing equipment automatically transmits the test results to the data acquisition and storage module, and passes the real-time test data to the data analysis and modeling module; S3: Data collection and storage stage: The data collection and storage module collects the test data of the sample in real time through the API interface, including the sample number, test time and toxin content; The collected data is classified and stored according to the sample number, sampling time and test date, and the data is backed up through a distributed database; S4: Data analysis and modeling stage: The data analysis and modeling module preprocesses the detection data received from the data acquisition and storage module, including data normalization, denoising, and outlier detection; The preprocessed data were used to train a machine learning model to establish a prediction model for DON residues; The data analysis and modeling module generates future prediction results of DON residues based on historical data and the latest test data, and passes the analysis results to the intelligent early warning and decision support module; S5: Intelligent early warning and decision support stage: The intelligent early warning and decision support module sets the early warning threshold according to the analysis results and prediction models generated by the data analysis and modeling module; When the detection data or prediction results exceed the set threshold, the warning generation unit triggers the warning signal and generates decision parameters for production adjustment or supply chain management through the data processing unit; Early warning and decision-making information is delivered to users through the visual display module; S6: Visualization display and report generation stage: The visualization display module obtains data from the data acquisition and storage module, the data analysis and modeling module, and the intelligent early warning and decision support module, and displays the test results, DON residue trends, and early warning information to the user in the form of charts or dashboards; Users can generate customized reports through the visualization module and choose to export them as PDF or Excel files; S7: Compliance and Report Generation Phase: The compliance and reporting module dynamically tracks updates to national and international food safety standards; The Compliance and Report Generation module automatically generates detailed test reports containing sample information, test results and analytical data, while recording audit information for each test step, including timestamps, operator information and equipment parameters.

Images