An artificial intelligence-based food safety detection management and control method and system

Abstract

The application discloses a food safety detection management and control method and system based on artificial intelligence, and relates to the technical field of safety detection management and control. In the system operation, by collecting various data related to the safety of additives in food, the concentration data of actual additives in food samples, abnormal detection data and data of additives under different environmental conditions are obtained to obtain a first data group, a second data group and a third data group. The original data collected are preprocessed, and after calculation, the food safety index Spaq is obtained. The safety of additives in the food production and processing process is monitored in real time, instant feedback and early warning are provided, a report is automatically generated based on the grade evaluation result of the real-time monitoring module, a treatment suggestion for unqualified additives is provided, and a corresponding alarm is sent. The detection algorithm and model are updated regularly to adapt to new technologies and standards, the detection accuracy and efficiency of the evaluation system are evaluated, and optimization and adjustment are performed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of safety detection and control, in particular to a food safety detection and control method and system based on artificial intelligence. BACKGROUND

[0002] Food safety issues have always been a global focus, and the use and management of food additives are particularly important. Food additives play a crucial role in food production, improving the taste, appearance, and shelf life of food. However, with the development of the food industry, the types and amounts of additives continue to increase, posing potential health risks, and the safety detection of additives is particularly critical because the consumer group of these foods is more sensitive to additives and has more stringent safety requirements. Therefore, safety detection of food additives is an important link to ensure food safety.

[0003] The application of artificial intelligence in the food safety detection and control system provides a new path to solve the many shortcomings of traditional detection methods. Traditional food safety detection often relies on manual operation, with low detection efficiency and difficulty in ensuring the accuracy and consistency of detection results. In particular, in the detection process of food additives, traditional detection methods are difficult to cope with complex ingredients and variable environmental conditions. In addition, traditional food safety detection systems lack real-time capability and cannot timely detect and warn potential safety risks.

[0004] The shortcomings of existing food safety detection systems mainly lie in the lack of timeliness and comprehensiveness of detection, resulting in potential food safety hazards that are not detected in time. For example, due to the limitations of manual detection, some subtle concentration changes and environmental factors may be overlooked, leading to problems such as non-compliance of additive concentration or insufficient stability. After these problems occur, it may lead to a decline in food quality and even have adverse effects on consumers' health. This situation cannot be monitored and handled in time during the production process, which may cause substandard products to flow into the market, ultimately leading to more serious food safety incidents. Therefore, there is an urgent need for an efficient, accurate, and comprehensive detection system to make up for the shortcomings of existing detection methods and ensure food safety. SUMMARY

[0005] To overcome the shortcomings of the prior art, the present application provides a food safety detection and control method and system based on artificial intelligence, which solves the problems mentioned in the background art.

[0006] To achieve the above purpose, the present application realizes the following technical scheme: a food safety detection and control system based on artificial intelligence, comprising a data acquisition module, a data preprocessing module, an intelligent detection module, a real-time monitoring module, a decision support module, and a system maintenance module.

[0007] The data acquisition module is used to collect various data related to the safety of additives in food, obtain concentration data of actual additives in food samples, abnormal detection data and data of additives under different environmental conditions, and obtain first, second and third data groups;

[0008] The data preprocessing module is used to clean, normalize and extract features from the collected raw data to ensure the accuracy and consistency of the data;

[0009] The intelligent detection module is used to classify and predict the type, concentration and safety risk of additives using machine learning algorithms, identify potential safety hazards using clustering algorithms and anomaly detection techniques, and train models based on historical data to calculate the food safety index Spaq;

[0010] The real-time monitoring module is used to monitor the safety of additives in food production and processing in real time, provide immediate feedback and early warning, display the current additive detection state, concentration and abnormal information, set threshold level evaluation, and timely alarm relevant personnel when exceeding or abnormal conditions are detected;

[0011] The decision support module is used to automatically generate reports containing additive concentration, stability evaluation and abnormal detection results based on the level evaluation results of the real-time monitoring module, provide suggestions for handling unqualified additives, including adjusting the formula, suspending production or recalling products, and issuing corresponding alarms;

[0012] The system maintenance module is used to periodically update detection algorithms and models to adapt to new technologies and standards, evaluate the detection accuracy and efficiency of the system, and optimize and adjust.

[0013] Preferably, the data acquisition module includes an additive concentration data acquisition unit, an abnormal detection data acquisition unit and an additive stability data acquisition unit;

[0014] The additive concentration data acquisition unit is used to collect concentration data of actual additives in food samples, collect and record actual additive concentrations, allowable concentration ranges and concentration deviations in food through high-performance liquid chromatographs, gas chromatographs and mass spectrometers, and obtain actual concentration data C act , allowable maximum concentration data C max , allowable minimum concentration data C min and concentration compliance rate R com , forming the first data group;

[0015] The abnormality detection data collection unit is used to collect abnormal situation data occurring in the food production and detection process. Through intelligent sensors, real-time monitoring systems and data analysis platforms, the abnormality detection rate, the number of abnormal types and the related data of system response are monitored and recorded in real time to obtain: abnormality detection rate data R anom , the number of abnormal type data N anom , detection sensitivity S sens , false alarm rate R false and abnormal response time data T resp , forming a second data set.

[0016] The additive stability data collection unit is used to collect the stability data of the additive under different environmental conditions. Through the accelerated aging test equipment, high-performance liquid chromatograph and environmental control system equipment, the degradation rate, stability test results and environmental impact factors of the additive are collected and recorded to obtain: degradation rate data R deg , stability test result data S test , environmental impact factor data E impact , shelf life P life and physical interaction value P interact , forming a third data set.

[0017] Preferably, the data preprocessing module includes a data cleaning unit and a feature extraction unit.

[0018] The data cleaning unit is used to clean the collected raw data, mainly including removing noise data, processing missing values and abnormal values.

[0019] The feature extraction unit is used to normalize the cleaned data to eliminate the differences between different data scales and extract key feature parameters to provide an optimized feature data set for subsequent algorithm modeling.

[0020] Preferably, the intelligent detection module includes a prediction unit.

[0021] The prediction unit is used to apply machine learning algorithms to classify and predict the type, concentration and safety risk of the additive to generate a food safety index Spaq, an additive concentration coefficient K c , a safety standard compliance coefficient K s and an additive stability coefficient K e .

[0022] Preferably, the real-time monitoring module includes a warning unit.

[0023] The early warning unit is used to monitor the safety of additives in food production and processing in real time. By comparing the food safety index Spaq with the first qualified threshold M and the second qualified threshold N, the detection state, concentration and abnormal information of the current additive are evaluated. Once the over-standard or abnormal situation is detected, the relevant personnel are immediately notified of the alarm to ensure that appropriate measures are taken in a timely manner.

[0024] Preferably, the corresponding measures include:

[0025] The food safety index Spaq is greater than or equal to the second qualified threshold N, indicating that the safety of food additives is more than 20% higher than the qualified state, and the existing production and processing process is continued, and regular monitoring and detection are carried out.

[0026] The first qualified threshold M is less than the food safety index Spaq, which is less than the second qualified threshold N, indicating that the safety of food additives is qualified, but there is still room for improvement. The monitoring frequency of concentration and stability is increased to discover factors affecting the food safety index Spaq in a timely manner.

[0027] The safety index Spaq is less than the first qualified threshold M, indicating that the safety of food additives has hidden dangers, and is determined to be unqualified. Production is immediately stopped, comprehensive production line inspection is carried out, the automatic early warning system is started, the relevant personnel are immediately notified, and the emergency plan is started.

[0028] Preferably, the decision support module includes a report generation unit and a processing suggestion unit.

[0029] The report generation unit is used to automatically generate a comprehensive report based on the level evaluation results of the real-time monitoring module. The report content includes additive concentration, stability evaluation and abnormal detection results, and provides trend analysis.

[0030] Preferably, the processing suggestion unit is used to automatically evaluate whether further action is needed based on the data provided by the report generation unit, including adjusting the additive formula, suspending production or recalling products. Once unqualified additives are detected, specific processing suggestions are generated, and corresponding alarms are sent through the system to notify relevant personnel to take action in a timely manner.

[0031] Preferably, the system maintenance module includes a model updating unit.

[0032] The model updating unit is used to periodically update and optimize the detection algorithm and model, evaluate the detection accuracy and efficiency of the system, identify areas for improvement, and make adjustments to adapt to new technologies and standards.

[0033] A food safety detection and control method based on artificial intelligence includes the following steps:

[0034] Step one: Collect data related to the safety of additives in food, obtain the concentration data of actual additives in food samples, abnormal detection data and data of additives under different environmental conditions, obtain the first, second and third data sets;

[0035] Step two: Clean, normalize and extract features from the collected raw data to ensure the accuracy and consistency of the data;

[0036] Step three: Use machine learning algorithms to classify and predict the type, concentration and safety risk of additives, apply clustering algorithms and anomaly detection techniques to identify potential safety hazards, train the model based on historical data, and calculate the food safety index Spaq;

[0037] Step four: Real-time monitoring of additive safety during food production and processing, providing immediate feedback and early warning, displaying the current additive detection status, concentration and abnormal information, setting threshold level evaluation, and issuing an alarm when exceeding the standard or abnormal situation is detected;

[0038] Step five: Based on the level evaluation results of the real-time monitoring module, automatically generate a report containing the concentration of additives, stability evaluation and abnormal detection results, provide suggestions for handling unqualified additives, including adjusting the formula, suspending production or recalling products, and issuing corresponding alarms;

[0039] Step six: Regularly update the detection algorithm and model to adapt to new technologies and standards, evaluate the detection accuracy and efficiency of the system, and optimize and adjust.

[0040] The present application provides a food safety detection and control method and system based on artificial intelligence, which has the following beneficial effects:

[0041] (1) When the system is running, by collecting data related to the safety of additives in food, obtaining the concentration data of actual additives in food samples, abnormal detection data and data of additives under different environmental conditions, obtaining the first, second and third data sets, preprocessing the collected raw data, calculating the food safety index Spaq, real-time monitoring of additive safety during food production and processing, providing immediate feedback and early warning, based on the level evaluation results of the real-time monitoring module, automatically generating a report, providing suggestions for handling unqualified additives, and issuing corresponding alarms, regularly updating the detection algorithm and model to adapt to new technologies and standards, evaluating the detection accuracy and efficiency of the system, and optimizing and adjusting.

[0042] (2) The food safety detection and control system based on artificial intelligence significantly improves the efficiency and accuracy of food additive safety detection through the coordinated work of six modules. The data collection module is responsible for comprehensively collecting the key additive concentration, anomaly detection and stability data in food samples, ensuring that all relevant factors are fully considered. The data preprocessing module cleans and extracts features from these data, eliminating noise and outliers in the data, laying a solid foundation for subsequent algorithm analysis. The intelligent detection module uses advanced machine learning algorithms to accurately classify and predict the type, concentration and potential safety risks of additives, achieving precise calculation of the food safety index Spaq.

[0043] (3) The real-time monitoring module monitors the safety of additives in the food production process in real time, and through the set threshold level evaluation, ensures that once the over-standard or abnormal situation is detected, it can timely issue an alarm and notify the relevant personnel. This real-time and automated monitoring greatly shortens the response time compared to traditional manual detection methods, improving the agility of detection. The decision support module generates comprehensive reports and proposes targeted handling suggestions based on the monitoring results, such as adjusting the formula, suspending production or recalling products, thereby further ensuring food safety. The system maintenance module updates the detection algorithm and model regularly to ensure that the system can continuously adapt to new technical standards and market demands, maintaining the efficiency and reliability of the detection system.

[0044] (4) Compared with traditional technical means, the introduction of this system not only improves the comprehensiveness and accuracy of food additive safety detection, but also significantly enhances the ability of real-time monitoring and emergency response. Traditional detection methods often rely on manual operation, which has the problems of low efficiency and being easily affected by human factors, while the automated and intelligent processing method of this system effectively solves these deficiencies. In addition, through continuous updating and optimization of algorithms, the system can keep up with the latest technological developments and industry standards, ensuring that food safety detection is always at the leading level. These improvements enable safety hazards in the food production process to be discovered and handled earlier, thereby effectively improving the overall level of food safety protection. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 is a block diagram of the food safety detection and control system based on artificial intelligence of the present application;

[0046] Figure 2 is a step schematic diagram of the food safety detection and control method based on artificial intelligence of the present application. DETAILED DESCRIPTION

[0047] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of the present application.

[0048] Embodiment 1

[0049] The present application provides an artificial intelligence-based food safety detection and control system, please refer to Figure 1 , comprising a data acquisition module, a data preprocessing module, an intelligent detection module, a real-time monitoring module, a decision support module and a system maintenance module.

[0050] The data acquisition module is used to collect various data related to the safety of additives in food, obtain the concentration data of actual additives in food samples, abnormal detection data and data of additives under different environmental conditions, obtain the first data set, the second data set and the third data set.

[0051] The data preprocessing module is used to clean, normalize and feature extract the collected raw data to ensure the accuracy and consistency of the data.

[0052] The intelligent detection module is used to classify and predict the type, concentration and safety risk of additives using machine learning algorithms, identify potential safety hazards using clustering algorithms and anomaly detection techniques, train the model based on historical data, and calculate the food safety index Spaq.

[0053] The real-time monitoring module is used to monitor the safety of additives in the production and processing of food in real time, provide immediate feedback and early warning, display the current additive detection state, concentration and abnormal information, set threshold level evaluation, and timely issue an alarm to inform relevant personnel when exceeding the standard or abnormal conditions are detected.

[0054] The decision support module is used to automatically generate a report containing the concentration of additives, stability evaluation and abnormal detection results based on the level evaluation results of the real-time monitoring module, provide suggestions for handling unqualified additives, including adjusting the formula, suspending production or recalling products, and issuing corresponding alarms.

[0055] The system maintenance module is used to periodically update the detection algorithm and model to adapt to new technologies and standards, evaluate the detection accuracy and efficiency of the system, and optimize and adjust.

[0056] In this embodiment, by collecting various data related to the safety of additives in food, obtaining the concentration data of actual additives in food samples, abnormal detection data and data of additives under different environmental conditions, obtaining the first data set, the second data set and the third data set, preprocessing the collected raw data, and calculating the food safety index Spaq, the real-time monitoring of the safety of additives in the production and processing of food, providing instant feedback and early warning, based on the level evaluation result of the real-time monitoring module, automatically generating a report containing the concentration of additives, stability evaluation and abnormal detection result, providing processing suggestions for unqualified additives, including adjusting the formula, suspending production or recalling products, and issuing corresponding alarms, regularly updating the detection algorithm and model to adapt to new technologies and standards, evaluating the detection accuracy and efficiency of the system, and optimizing and adjusting.

[0057] Embodiment 2

[0058] This embodiment is an explanation and illustration in embodiment 1, please refer to Figure 1 , specifically: the data acquisition module includes an additive concentration data acquisition unit, an abnormal detection data acquisition unit and an additive stability data acquisition unit;

[0059] The additive concentration data acquisition unit is used to acquire the concentration data of actual additives in food samples, through high performance liquid chromatograph, gas chromatograph and mass spectrometer equipment, to collect and record the actual concentration of additives in food, the allowable concentration range and the concentration deviation, to obtain: actual concentration data C act , allowable maximum concentration data C max , allowable minimum concentration data C min and concentration compliance rate R com , forming the first data set;

[0060] The abnormal detection data acquisition unit is used to acquire the abnormal situation data occurred in the production and detection process of food, through intelligent sensor, real-time monitoring system and data analysis platform, to real-time monitor and record the abnormal detection rate, the number of abnormal types and the related data of system response, to obtain: abnormal detection rate data R anom , the number of abnormal type data N anom , detection sensitivity S sens , false alarm rate R false and abnormal response time data T resp , forming the second data set;

[0061] The additive stability data acquisition unit is used to acquire the stability data of additives under different environmental conditions, through accelerated aging test equipment, high performance liquid chromatograph and environmental control system equipment, to collect and record the degradation rate of additives, stability test results and environmental impact factors, to obtain: degradation rate data Rdeg , stability test result data S test , environmental impact factor data E impact , shelf life P life and physical interaction value P interact , forming a third data set.

[0062] The data preprocessing module includes a data cleaning unit and a feature extraction unit;

[0063] The data cleaning unit is used to clean the collected raw data, mainly including removing noise data, processing missing values and abnormal values;

[0064] The feature extraction unit is used to normalize the cleaned data to eliminate the differences between different data scales and extract key feature parameters to provide an optimized feature data set for subsequent algorithm modeling.

[0065] In this embodiment, the data acquisition module can comprehensively and accurately collect key data related to the safety of food additives through the additive concentration data acquisition unit, the anomaly detection data acquisition unit and the additive stability data acquisition unit. The actual concentration data obtained by the high-performance liquid chromatograph, gas chromatograph and mass spectrometer equipment and the anomaly detection data collected by the intelligent sensor and real-time monitoring system enable the system to accurately assess the safety of additives. This comprehensive data acquisition improves the system's ability to identify food safety hazards and provides a solid foundation for subsequent detection and analysis. The data preprocessing module effectively removes noise and abnormal values in the raw data through the data cleaning unit and the feature extraction unit, and normalizes the data. This preprocessing ensures the accuracy and consistency of the data and avoids analysis bias caused by data quality problems. By extracting key feature parameters, this module also provides an optimized feature data set for subsequent algorithm modeling, improving the efficiency and accuracy of intelligent detection algorithms and ensuring the reliability of the final detection results.

[0066] Embodiment 3

[0067] This embodiment is an explanation and description in embodiment 1, please refer to Figure 1 , specifically: the intelligent detection module includes a prediction unit;

[0068] The prediction unit is used to apply machine learning algorithms to classify and predict the type, concentration and safety risk of additives, generate food safety index Spaq, additive concentration coefficient K c , safety standard compliance coefficient K s and additive stability coefficient K e ;

[0069] The food safety index Spaq is calculated by the following formula:

[0070] Spaq = a * K c + b * K s + g * K e ;

[0071] In the formula, K c represents the additive concentration coefficient, K s represents the safety standard compliance coefficient, and K e represents the additive stability coefficient, a, b, and g respectively represent the proportional coefficients of the additive concentration coefficient K c , the safety standard compliance coefficient K s , and the additive stability coefficient K e ;

[0072] The additive concentration coefficient K c is calculated by the following formula:

[0073]

[0074] In the formula, C act represents the actual concentration data, C max represents the maximum allowable concentration data, C min represents the minimum allowable concentration data, and R com represents the concentration compliance rate.

[0075] The concentration compliance rate R com is calculated by the following formula:

[0076]

[0077] The safety standard compliance coefficient K s is calculated by the following formula:

[0078]

[0079] In the formula, R anom represents the anomaly detection rate data, N anom represents the number of anomaly type data, S sens represents the detection sensitivity, R false represents the false positive rate, and T resp represents the anomaly response time data.

[0080] The anomaly detection rate data R anom is calculated by the following formula:

[0081] The false positive rate R false is calculated by the following formula:

[0082] The additive stability coefficient K e The acquisition is calculated by the following formula:

[0083]

[0084] In the formula, R deg represents the degradation rate data, S test represents the stability test result data, E impact represents the environmental impact factor data, P life represents the shelf life, P interact represents the physical interaction value.

[0085] The degradation rate data R deg is calculated by the following formula:

[0086] In this embodiment, the intelligent detection module uses machine learning algorithms to classify and predict the type, concentration and safety risk of additives through the prediction unit. The generated food safety index Spaq, additive concentration coefficient K c , safety standard compliance coefficient K s and additive stability coefficient K e can fully reflect the safety status of additives in food. This comprehensive evaluation method based on multi-dimensional data is more accurate than traditional single parameter detection, significantly improving the prediction and management ability of food safety risk. The system calculates the additive concentration coefficient K c , safety standard compliance coefficient K s and additive stability coefficient K e through a series of complex algorithm formulas, which can quickly draw conclusions. These algorithms not only can process a large amount of data in real time, but also can improve detection accuracy by continuously optimizing the model. This efficient and accurate algorithm-driven detection method significantly shortens the detection cycle, enabling food production and supervision to respond in a shorter time, thereby improving the overall efficiency of food safety management.

[0087] Embodiment 4

[0088] This embodiment is an explanation and description in embodiment 1, please refer to Figure 1 , specifically: the real-time monitoring module includes a warning unit;

[0089] The early warning unit is used to monitor the safety of additives in food production and processing in real time. By comparing the food safety index Spaq with the first qualified threshold M and the second qualified threshold N, the detection state, concentration and abnormal information of the current additive are evaluated. Once the over-standard or abnormal situation is detected, an alarm is immediately sent to the relevant personnel to ensure that appropriate measures are taken in a timely manner.

[0090] The food safety index Spaq is greater than the second qualified threshold N, indicating that the safety of food additives is more than 20% higher than the qualified state. The existing production and processing process is continued, regular monitoring and detection are carried out, and the concentration of additives is ensured to be more stable between Cmax and Cmin by further optimizing the automatic control of the additive feeding system. More precise flow control valves or real-time concentration monitoring systems can be installed, and the production environment is regularly evaluated and adjusted, such as controlling temperature, humidity and other environmental factors, to further improve the stability Stest of additives.

[0091] The first qualified threshold M is less than the food safety index Spaq, which is less than the second qualified threshold N, indicating that the safety of food additives is qualified, but there is still room for improvement. The monitoring frequency of concentration and stability is increased to timely discover factors affecting the food safety index Spaq, and each batch of production is monitored and recorded in more detail, especially the optimization of key process parameters such as temperature and pressure to ensure the stability of each batch of products. The storage conditions of additives are gradually improved during production and storage, such as controlling light, oxygen content, etc., to reduce the degradation rate Rdeg.

[0092] The safety index Spaq is less than the first qualified threshold M, indicating that the safety of food additives has hidden dangers and is unqualified. Production is immediately suspended, the production line is thoroughly investigated, and the equipment or raw materials that appear problems are adjusted or replaced. The concentration control system is recalibrated, the automatic early warning system is started, the relevant personnel are immediately notified, and the emergency plan is started.

[0093] The decision support module includes a report generation unit and a processing suggestion unit.

[0094] The report generation unit is used to automatically generate a comprehensive report based on the level evaluation results of the real-time monitoring module. The report content includes additive concentration, stability evaluation and abnormal detection results, and provides trend analysis.

[0095] The processing suggestion unit is used to automatically evaluate whether further action is needed based on the data provided by the report generation unit, including adjusting the additive formula, suspending production or recalling products. Once unqualified additives are detected, specific processing suggestions are generated and corresponding alarms are sent through the system to notify relevant personnel to take action in a timely manner.

[0096] The system maintenance module includes a model updating unit;

[0097] The model updating unit is used to periodically update and optimize the detection algorithms and models, evaluate the detection accuracy and efficiency of the system, identify areas for improvement, and make adjustments to adapt to new technologies and standards.

[0098] In this embodiment, the early warning unit in the real-time monitoring module achieves comprehensive dynamic monitoring of the food production process by comparing the food safety index Spaq with the preset threshold in real time. When the safety of additives is lower than the early warning value, the system will immediately issue an alarm to notify relevant personnel to take emergency measures quickly. This instant early warning and response mechanism effectively prevents potential food safety hazards, provides stronger protection for food safety, and significantly reduces the probability of accidents. Through the decision support module, the system can automatically generate detailed monitoring reports and provide specific handling recommendations based on the test results. Whether it is adjusting the formula, suspending production, or recalling products, the system can provide the best action plan. This intelligent decision support not only improves the efficiency of food safety management, but also makes the production process more precise and controllable, ensuring the high quality and safety of the final product.

[0099] Embodiment 5

[0100] A food safety detection and control method based on artificial intelligence, please refer to Figure 2 , specifically: including the following steps:

[0101] Step 1: Collect data related to the safety of additives in food, obtain the concentration data of actual additives in food samples, abnormal detection data and data of additives under different environmental conditions, obtain the first data set, the second data set and the third data set;

[0102] Step 2: Clean, normalize and feature extract the collected raw data to ensure the accuracy and consistency of the data;

[0103] Step 3: Use machine learning algorithms to classify and predict the type, concentration and safety risk of additives, apply clustering algorithms and anomaly detection techniques to identify potential safety hazards, train the model based on historical data, and calculate the food safety index Spaq;

[0104] Step 4: Real-time monitoring of additive safety during food production and processing, providing instant feedback and early warning, displaying the current additive detection status, concentration and abnormal information, setting threshold level evaluation, and issuing an alarm to notify relevant personnel when exceeding the standard or detecting abnormal conditions;

[0105] Step five: Based on the ranking evaluation results of the real-time monitoring module, automatically generate a report containing the additive concentration, stability evaluation and abnormal detection results, provide processing suggestions for unqualified additives, including adjusting the formula, suspending production or recalling products, and send corresponding alerts;

[0106] Step six: Regularly update the detection algorithm and model to adapt to new technologies and standards, evaluate the detection accuracy and efficiency of the system, and optimize and adjust.

[0107] In this embodiment, in the first and second steps, through systematic data collection and preprocessing, all key data related to the safety of additives in food samples are accurately recorded and processed. The concentration data, abnormal detection data and stability data of additives under different environmental conditions obtained through various advanced devices build a detailed and reliable data foundation. The subsequent cleaning and normalization processing further improves the accuracy and consistency of the data, laying a solid foundation for the intelligent analysis of the system. In the third to fourth steps, the system uses machine learning algorithms and clustering techniques to accurately classify and predict the type, concentration and potential safety risks of additives. The real-time monitoring module calculates the food safety index Spaq and sets the threshold, realizing dynamic monitoring and risk management of the whole food production process. When abnormal or excessive conditions are detected, the system can immediately issue a warning and notify relevant personnel, ensuring that food safety hazards are identified and addressed in a timely manner. In the fifth and sixth steps, the system not only generates detailed detection reports based on real-time monitoring data, but also provides specific handling suggestions based on the reports to ensure that every step in the production process meets safety standards. In addition, the system has a model updating function that can continuously optimize the detection algorithm and model to adapt to new technologies and industry standards. This continuous optimization mechanism enables the system to maintain efficient and accurate detection capabilities in the long run, improving the overall safety and product quality of food production.

[0108] Although embodiments of the present application have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and alterations can be made thereto without departing from the principles and spirit of the present application, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An artificial intelligence-based food safety detection management and control system, characterized in that: The system comprises a data collection module, a data preprocessing module, an intelligent detection module, a real-time monitoring module, a decision support module, and a system maintenance module. The data collection module is used to collect various data related to the safety of additives in food, obtain concentration data of actual additives in food samples, abnormal detection data, and data of additives under different environmental conditions, and obtain first, second, and third data groups. The data preprocessing module is used to clean, normalize, and extract features from the collected raw data to ensure the accuracy and consistency of the data. The intelligent detection module is used to classify and predict the types, concentrations, and safety risks of additives using machine learning algorithms, identify potential safety hazards using clustering algorithms and anomaly detection techniques, and train models based on historical data to calculate the food safety index Spaq. The real-time monitoring module is used to monitor the safety of additives in food production and processing in real time, provide immediate feedback and warnings, display the current detection status, concentration, and abnormal information of additives, set threshold levels for evaluation, and issue alerts to relevant personnel when over-standard or abnormal conditions are detected. The decision support module is used to automatically generate reports containing additive concentration, stability evaluation, and abnormal detection results based on the evaluation results of the real-time monitoring module, provide suggestions for handling unqualified additives, including adjusting formulations, suspending production, or recalling products, and issue corresponding alerts. The system maintenance module is used to periodically update detection algorithms and models to adapt to new technologies and standards, evaluate the detection accuracy and efficiency of the system, and optimize and adjust the system. The real-time monitoring module comprises a warning unit. The warning unit is used to monitor the safety of additives in food production and processing in real-time, compare the food safety index Spaq with the pre-set first and second qualified thresholds M and N, evaluate the current detection status, concentration, and abnormal information of additives, and immediately issue alerts to relevant personnel when over-standard or abnormal conditions are detected to ensure timely measures are taken. The corresponding measures include: When the food safety index Spaq is greater than or equal to the second qualified threshold N, it indicates that the safety of food additives is more than 20% higher than the qualified state, and the existing production and processing process can be maintained, with regular monitoring and detection. When the first qualified threshold M is less than the food safety index Spaq and less than the second qualified threshold N, it indicates that the safety of food additives is qualified but still has room for improvement, and the monitoring frequency of concentration and stability should be increased to timely identify factors affecting the food safety index Spaq. When the safety index Spaq is less than the first qualified threshold M, it indicates that the safety of food additives has potential risks, and the production should be stopped immediately, a comprehensive production line inspection should be conducted, the automatic warning system should be started, relevant personnel should be notified immediately, and an emergency plan should be activated.

2. The food safety detection and management system based on artificial intelligence according to claim 1, characterized in that: The data collection module comprises an additive concentration data collection unit, an abnormal detection data collection unit, and an additive stability data collection unit. The additive concentration data acquisition unit is used to acquire the concentration data of the actual additive in the food sample. Through high-performance liquid chromatograph, gas chromatograph and mass spectrometer equipment, the actual concentration, allowable concentration range and concentration deviation of the additive in the food are collected and recorded to obtain actual concentration data Cact, allowable maximum concentration data Cmax, allowable minimum concentration data Cmin and concentration compliance rate Rcom, forming a first data group. The anomaly detection data acquisition unit is used to acquire anomaly situation data occurring in the food production and detection process. Through intelligent sensors, real-time monitoring systems and data analysis platforms, the anomaly detection rate, anomaly type quantity and system response related data are monitored and recorded in real time to obtain anomaly detection rate data Ranom, anomaly type quantity data Nanom, detection sensitivity Ssens, false alarm rate Rfalse and anomaly response time data Tresp, forming a second data group. The additive stability data acquisition unit is used to acquire the stability data of the additive under different environmental conditions. Through accelerated aging test equipment, high-performance liquid chromatograph and environmental control system equipment, the degradation rate, stability test results and environmental impact factors of the additive are collected and recorded to obtain degradation rate data Rdeg, stability test result data Stest, environmental impact factor data Eimpact, shelf life Plife and physical interaction value Pinteract, forming a third data group.

3. The food safety detection and management system based on artificial intelligence according to claim 1, characterized in that: The data preprocessing module includes a data cleaning unit and a feature extraction unit. The data cleaning unit is used to clean the collected raw data, mainly including removing noise data, processing missing values and abnormal values. The feature extraction unit is used to normalize the cleaned data to eliminate the differences between different data scales and extract key feature parameters to provide an optimized feature data set for subsequent algorithm modeling.

4. The food safety detection and management system based on artificial intelligence according to claim 1, characterized in that: The intelligent detection module includes a prediction unit. The prediction unit is used to apply machine learning algorithms to classify and predict the type, concentration and safety risk of the additive to generate a food safety index Spaq, an additive concentration coefficient Kc, a safety standard compliance coefficient Ks and an additive stability coefficient Ke.

5. The food safety detection and management system based on artificial intelligence according to claim 1, characterized in that: The decision support module includes a report generation unit and a processing suggestion unit. The report generation unit is used to automatically generate a comprehensive report based on the level evaluation results of the real-time monitoring module. The report content includes additive concentration, stability evaluation and anomaly detection results, and provides trend analysis.

6. The food safety detection and control system based on artificial intelligence according to claim 5, characterized in that: The processing suggestion unit is used to automatically evaluate whether further action needs to be taken based on the data provided by the report generation unit, including adjusting the additive formula, suspending production or recalling the product. Once a non-compliant additive is detected, specific processing suggestions are generated immediately, and corresponding alarms are sent through the system to notify relevant personnel to take timely action.

7. The food safety detection and management system based on artificial intelligence according to claim 1, characterized in that: The system maintenance module includes a model updating unit. The model updating unit is used to periodically update and optimize the detection algorithm and model, evaluate the detection accuracy and efficiency of the system, identify areas for improvement, and make adjustments to adapt to new technologies and standards.

8. An artificial intelligence-based food safety detection management and control method applied to the artificial intelligence-based food safety detection management and control system of any one of claims 1-7, characterized in that: It includes the following steps: Step one: Collect data related to the safety of additives in food, obtain the concentration data of actual additives in food samples, abnormal detection data and data of additives under different environmental conditions, obtain the first, second and third data sets; Step two: Clean, normalize and extract features from the collected raw data to ensure data accuracy and consistency; Step three: Use machine learning algorithms to classify and predict additive types, concentrations and safety risks, apply clustering algorithms and anomaly detection techniques to identify potential safety hazards, train the model based on historical data, and calculate the food safety index Spaq; Step four: Real-time monitoring of additive safety during food production and processing, providing immediate feedback and early warning, displaying current additive detection status, concentration and abnormal information, setting threshold level evaluation, and issuing alarm notifications to relevant personnel when exceeding the standard or detecting abnormal conditions; Step five: Based on the level evaluation results of the real-time monitoring module, automatically generate a report containing additive concentration, stability evaluation and abnormal detection results, provide suggestions for handling unqualified additives, including adjusting the formula, suspending production or recalling products, and issuing corresponding alerts; Step six: Regularly update the detection algorithm and model to adapt to new technologies and standards, evaluate the detection accuracy and efficiency of the system, and optimize and adjust.

Images