Agricultural food product quality safety monitoring and standardization system based on technical trade measures

Abstract

The invention provides an agricultural food product quality safety monitoring and standardization system based on technical trade measures. The agricultural food product quality safety monitoring and standardization system comprises a data acquisition module, a data processing module, a model training and risk assessment module and a standardization execution module, the data acquisition module acquires data related to agricultural food product quality safety from a plurality of data sources, the data processing module cleans, converts and integrates the data, and the model training and risk assessment module assesses and predicts quality safety risks based on preprocessed data. And the standardization execution module formulates and executes standardization measures according to the risk assessment result. According to the system, through comprehensive data acquisition, efficient data processing, dynamic risk assessment and timely standardization measures, the efficiency and accuracy of agricultural food product quality safety monitoring are improved, the model performance is continuously improved through a feedback and optimization mechanism, and it is ensured that products meet the requirements of technical trade measures.

Description

Technical Field

[0001] The present invention belongs to the technical field of standardization management, and particularly relates to a quality safety monitoring and standardization system for agricultural and food products based on technical trade measures. Background Art

[0002] With the acceleration of the globalization process, the international trade of agricultural and food products has become increasingly frequent. Technical Barriers to Trade (TBT) are a series of technical regulations, standards, and conformity assessment procedures adopted by various countries to protect human health, the safety of animal and plant life, and the environment. While these measures ensure the quality and safety of products, they also have an important impact on international trade.

[0003] However, there are still some deficiencies in the existing quality safety monitoring systems for agricultural and food products in dealing with technical trade measures. Currently, the quality safety monitoring systems for agricultural and food products mainly rely on traditional detection methods and manual management methods. Existing systems often only focus on data in a certain aspect, such as production process data or market detection data, lacking the comprehensive collection and integration of technical trade measure data. Traditional methods rely on manual data processing, which is inefficient and error-prone, and difficult to meet the processing requirements of large-scale data. Existing systems mostly adopt static assessment methods, unable to dynamically reflect the quality safety risks of agricultural and food products at different time periods, resulting in inaccurate assessment results. Due to the lack of an effective risk assessment mechanism, existing systems are difficult to formulate and implement standardization measures in a timely manner and cannot quickly respond to quality safety problems.

[0004] The existing quality safety monitoring systems for agricultural products mainly rely on laboratory testing and on-site inspections, with low data collection and processing efficiency, unable to meet the needs of modern agricultural product trade. Although Internet of Things technology has certain applications in the traceability of agricultural product quality safety, there are still deficiencies in dealing with technical trade measures, lacking the integration of multi-source data and the ability of dynamic risk assessment.

[0005] In view of the above deficiencies of the existing technologies, the present application proposes a quality safety monitoring and standardization system for agricultural and food products based on technical trade measures. Summary of the Invention

[0006] The purpose of the present invention is to provide a quality safety monitoring and standardization system for agricultural and food products based on technical trade measures to improve the deficiencies of the existing technologies. The technical solution of the present invention is as follows: A quality safety monitoring and standardization system for agricultural and food products based on technical trade measures, the system includes a data collection module, a data processing module, a model training and risk assessment module, and a standardization execution module that are connected in sequence.

[0007] Further, the data acquisition module is used to obtain basic data from multiple data sources. The basic data includes technical trade measure data related to the quality and safety of agricultural and food products and the production control sequence data of the agricultural and food products. The technical trade measure data at least includes product standards, inspection and quarantine requirements, certification systems, and label and marking regulations; and sends the collected basic data to the data processing module.

[0008] Further, the data preprocessing module is used to clean, transform, and integrate the received basic data to form preprocessed data with a unified data format, and send it to the model training and risk assessment module.

[0009] Further, the model training and risk assessment module is used to conduct quality and safety risk assessment on agricultural and food products based on the preprocessed data. The model training and risk assessment module is also used to train a prediction model based on the preprocessed data and quantitatively predict the risk of agricultural and food products at the next moment based on the prediction model.

[0010] Further, the standardization execution module is used to formulate and execute corresponding standardization measures according to the risk assessment results.

[0011] Further, the data acquisition module at least includes: a web crawler sub-module, a data interface sub-module, and a user input sub-module.

[0012] The web crawler sub-module is used to crawl public technical trade measure data from the Internet; the data interface sub-module is used to dock with the data systems of production enterprises, government departments, and industry associations to obtain the production control sequence data of agricultural and food products and proprietary data related to technical trade measures; the user input sub-module is used to collect data manually input by users.

[0013] Further, the model training and risk assessment module includes a risk assessment model based on multivariate time series analysis. The model uses the following formula for risk assessment: Where, represents the risk assessment value at time , is the constant term of the model, , is the regression coefficient of each variable , is the number of input variables, represents the input variables at time , including technical trade measure data and production control sequence data; is the error term at time , which follows a normal distribution.

[0014] In addition, the model also includes an autoregressive moving average sub-model for capturing the autocorrelation of the data: where are the coefficients of the autoregressive part; are the coefficients of the moving average part; and are the orders of autoregression and moving average respectively.

[0015] Furthermore, the model training and risk assessment module includes: a data division sub-module, a feature selection sub-module, and a model evaluation sub-module.

[0016] The data division sub-module is used to divide the preprocessed data into a training set and a test set for model training and validation.

[0017] The feature selection sub-module is used to select the feature variables that have a significant impact on the quality and safety risk assessment of agro-food products from the preprocessed data.

[0018] The model evaluation sub-module is used to evaluate the trained model with the test set data, calculate evaluation indicators such as the accuracy, recall rate, and F1 score of the model to determine the performance and reliability of the model.

[0019] The model training and risk assessment module is trained based on the training set data using multivariate time series analysis and the autoregressive moving average sub-model, and the model is optimized and adjusted according to the evaluation indicators of the model evaluation sub-module.

[0020] Furthermore, the quantitative prediction of the risk of agro-food products at the next moment based on the prediction model includes: extracting time series data from the preprocessed data, inputting the time series data using the trained prediction model; calculating the risk assessment value ; outputting the risk assessment value as the quantitative prediction result of the risk at the next moment.

[0021] Furthermore, the standardization execution module formulates standardization measures based on fuzzy logic reasoning, and the fuzzy logic reasoning is expressed as: where represents the output value of the standardization measure, represents the input variable corresponding fuzzy membership function; represents the output value of each fuzzy rule.

[0022] The rule used in the fuzzy logic reasoning is expressed as: ; where and represent input variables; and represent the fuzzy sets of the input variables; represents the fuzzy set of the output variable.

[0023] Furthermore, the standardization execution module is also used to feedback the effect of the executed standardization measures to the model training and risk assessment module, and the model training and risk assessment module adjusts and optimizes the model according to the feedback information, regularly updates the data and retrains the model.

[0024] Furthermore, the feature selection sub-module selects feature variables that have a significant impact on the quality and safety risk assessment of agro-food products from the preprocessed data through feature selection, calculates the correlation between each feature variable and the target variable, and selects the feature variables with a significant impact.

[0025] The formula for feature selection is expressed as: where represents the correlation between the feature variable and the target variable, represents the feature variable, represents the target variable, and respectively represent the means of the feature variable and the target variable, is the number of samples.

[0026] The beneficial effects of the present invention are as follows: By comprehensively obtaining technical trade measure data and production control sequence data, the present invention uses a data preprocessing module to clean, transform and integrate the collected data to form preprocessed data in a unified format. Based on multivariate time series analysis and autoregressive moving average models, it dynamically conducts quality and safety risk assessment on agro-food products. Through fuzzy logic reasoning, corresponding standardization measures are formulated and implemented according to the risk assessment results, quickly responding to quality and safety issues, ensuring that the products meet the requirements of technical trade measures, and providing technical support for relevant enterprises to cope with technical trade measures. Description of the Drawings

[0027] Figure 1 is a schematic diagram of the composition of the quality and safety monitoring and standardization system for agro-food products based on technical trade measures of the present invention. Detailed Embodiments

[0028] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0029] As Figure 1 shown, it is a schematic diagram of the composition of a quality safety monitoring and standardization system for agricultural and food products based on technical trade measures. The system includes a data acquisition module, a data processing module, a model training and risk assessment module, and a standardization execution module that are connected in sequence.

[0030] The data acquisition module is used to obtain basic data from multiple data sources. The basic data includes technical trade measure data related to the quality safety of agricultural and food products and the production control sequence data of the agricultural and food products. The technical trade measure data at least includes product standards, inspection and quarantine requirements, certification systems, and label and marking regulations; and sends the collected basic data to the data processing module.

[0031] The data preprocessing module is used to clean, transform, and integrate the received basic data to form preprocessed data with a unified data format, and send it to the model training and risk assessment module.

[0032] When the data processing module processes the received basic data, it first performs data cleaning, including removing outliers, filling in missing values, and eliminating duplicate data. For example, for abnormally high values in pesticide use records, the system uses the moving median method for correction; for missing test data, it is interpolated and filled according to data at adjacent time points. Data transformation includes unit unification, format standardization, and numerical normalization processing, such as uniformly converting pesticide residue limit standards in different countries to the unit of mg / kg. Data integration associates data from different sources according to dimensions such as product categories, time series, and quality indicators to form a structured data set, providing a basis for subsequent modeling.

[0033] The model training and risk assessment module is used to perform quality safety risk assessment on agricultural and food products based on the preprocessed data. The model training and risk assessment module is also used to train a prediction model based on the preprocessed data and quantitatively predict the risk of agricultural and food products at the next moment based on the prediction model.

[0034] The standardization execution module is used to formulate and execute corresponding standardization measures according to the risk assessment results.

[0035] The data acquisition module at least includes: a web crawler sub-module, a data interface sub-module, and a user input sub-module.

[0036] The web crawler sub-module is used to crawl public technical trade measure data from the Internet; the data interface sub-module is used to dock with the data systems of production enterprises, government departments and industry associations to obtain production control sequence data of agricultural and food products and proprietary data related to technical trade measures; the user input sub-module is used to collect data manually input by users. The web crawler sub-module adopts a distributed crawler architecture and automatically crawls the latest technical trade measure information from more than 50 official websites such as the World Trade Organization (WTO), the Codex Alimentarius Commission (CAC), and the European Food Safety Authority (EFSA) every day. The system uses natural language processing technology to process the crawled unstructured text and identify key limit standards, inspection methods and compliance requirements. The data interface sub-module supports two interface methods, REST API and SOAP, and securely docks with the enterprise ERP system and the government supervision platform to synchronize information such as production process data, inspection reports and certification records in real time, and the average data transmission delay is less than 5 seconds.

[0037] Taking the quality safety monitoring and standardization of apples produced by a certain agricultural and food product formula as an example, the web crawler sub-module crawls technical trade measure data on apples from the Internet, including product standards, inspection and quarantine requirements, certification systems and label regulations; the data interface sub-module docks with the data systems of apple production enterprises, government departments and industry associations to obtain the production control sequence data of apples; the user input sub-module collects data manually input by users, such as farmers' production logs, field management records, etc.

[0038] Technical trade measure data includes: product standards, such as the size, color, sugar content, etc. of apples; inspection and quarantine requirements, such as pesticide residue limits, heavy metal limits, etc.; certification systems, such as organic certification, green food certification, etc.; label regulations, such as place of origin, production date, shelf life, etc. Production control sequence data includes: fertilization records, pesticide use records, irrigation records, picking records, etc.

[0039] The model training and risk assessment module includes a risk assessment model based on multivariate time series analysis, and the model uses the following formula for risk assessment: Where, represents the risk assessment value at time , is the constant term of the model, , is the regression coefficient of each variable , is the number of input variables, represents time Input variables at a given time, including technical trade measure data and production control sequence data; is time The error term at a given time, which follows a normal distribution.

[0040] In addition, the model also includes an autoregressive moving average sub-model for capturing the autocorrelation of the data: where, are the coefficients of the autoregressive part; are the coefficients of the moving average part; and are the orders of autoregression and moving average respectively.

[0041] The model training and risk assessment module includes: a data partitioning sub-module, a feature selection sub-module, and a model evaluation sub-module.

[0042] The data partitioning sub-module is used to partition the preprocessed data into a training set and a test set for model training and validation.

[0043] The feature selection sub-module is used to select feature variables that have a significant impact on the quality and safety risk assessment of agro-food products from the preprocessed data; in practical applications, the feature selection sub-module calculates the correlation between different feature variables and the quality and safety risk, and screens out the key variables with greater influence.

[0044] For example, for fruit products, the correlation between feature variables such as pesticide residues, heavy metal content, and microbial indicators and the quality and safety risk is usually high (correlation coefficient > 0.7), and they will be preferentially selected; while for some specific products, such as imported wine, the correlation of feature variables such as sulfur dioxide content and alcohol content is also significant (correlation coefficient is about 0.65). The system usually selects feature variables with an absolute value of the correlation coefficient greater than 0.4 as model inputs to ensure the accuracy and computational efficiency of the model.

[0045] The model evaluation sub-module is used to evaluate the trained model with the test set data, and calculate evaluation indicators such as the accuracy, recall rate, and F1 score of the model to determine the performance and reliability of the model.

[0046] The model evaluation sub-module uses the cross-validation method to evaluate the model performance, and specifically calculates indicators such as Precision, Recall, and F1 score. For quality and safety risk assessment, taking apple products as an example, the system divides the risk assessment value R into three levels: low risk (R < 0.3), medium risk (0.3 ≤ R < 0.7), and high risk (R ≥ 0.7). On the test dataset, the recognition precision of the model for high-risk apple samples reaches 92%, the recall rate is 88%, and the F1 score is 0.90; for medium-risk and low-risk samples, the F1 scores are 0.85 and 0.88 respectively. When the overall accuracy of the model is lower than 85%, the system will automatically adjust the feature variable selection or model parameters and retrain.

[0047] The model training and risk assessment module is based on the training set data, uses multivariate time series analysis and autoregressive moving average sub-models for training, and optimizes and adjusts the model according to the evaluation indicators of the model evaluation sub-module.

[0048] The model training process adopts an iterative optimization strategy. In the initial stage, a larger learning rate (0.01) is used for rapid convergence, and the learning rate is reduced (0.001) in the later stage for fine-tuning. The training adopts a batch processing method, with each batch containing 128 samples, and the training iteration times are set to 1000 times or until the change rate of the loss function is less than 0.1%. For different types of agricultural and food products, the system will automatically adjust the orders of p and q in the model. For example, for fruit products with obvious seasonality, p is usually set to 12 (corresponding to the annual cycle), while for vegetable products with a shorter production cycle, p may be set to 4 - 6. After the model training is completed, the system records the key parameter values. For a specific product, the constant term α of the model is 0.15, the range of the regression coefficient β value is from 0.21 to 0.45, and the range of the autoregressive part coefficient φ value is from 0.18 to 0.32.

[0049] The standardization execution module automatically generates corresponding standardization measure suggestions according to the risk assessment results; when the system detects that the pesticide residue risk assessment value of a batch of apple products exceeds the warning threshold (R > 0.65), the following standardization measures will be triggered: adjusting the safety interval before harvesting, increasing the detection frequency of specific pesticides, improving the product traceability code information, and revising the internal quality control standards. In addition, the system will also automatically generate targeted improvement plans according to the risk type and degree, such as specific suggestions like "reducing the usage amount of organophosphorus pesticides such as chlorpyrifos and methamidophos" and "extending the interval between pesticide application and harvesting by at least 15 days" to help production enterprises meet the requirements of technical trade measures in the export target country.

[0050] Quantitatively predicting the risk of agro-food products at the next moment based on the prediction model includes: extracting time series data from the preprocessed data and inputting the time series data into the trained prediction model; calculating the risk assessment value at the next moment ; outputting the risk assessment value , as the quantitative risk prediction result at the next moment.

[0051] For example, there is the following data: input variables at time t , ; regression coefficients , ; constant term ; error term . Calculate the risk assessment value at the current moment : ; ; ; Calculate the risk assessment value at the next moment : Assume that the input variables at the next moment , ; ; ; .

[0052] The standardization execution module formulates standardization measures based on fuzzy logic reasoning, and the fuzzy logic reasoning is expressed as: where represents the output value of the standardization measure, represents the fuzzy membership function corresponding to the input variable ; represents the output value of each fuzzy rule.

[0053] The rules used in the fuzzy logic reasoning are expressed as: ; where and represent input variables; and represent the fuzzy sets of the input variables; represents the fuzzy set of the output variable.

[0054] For example, IF the pesticide residue isLow AND the fertilization amount isOptimal THEN the risk isLow; IF pesticide residue isHigh AND fertilization amount isExcessive THEN risk isHigh.

[0055] The standardized execution module is also used to feedback the effect of the executed standardized measures to the model training and risk assessment module, and the model training and risk assessment module adjusts and optimizes the model according to the feedback information, regularly updates the data and retrains the model.

[0056] The standardized execution module adopts a closed-loop feedback mechanism to track and evaluate the effect of the executed standardized measures. The system sets multiple key performance indicators (KPIs), such as the improvement rate of the qualified rate, the decline rate of the risk value, the completion degree of standard execution, etc., and establishes an evaluation period (usually 7 days or 14 days). After each evaluation period ends, the system automatically generates a performance report and calculates the effectiveness index E of the measure = Σ(wi × KPIi), where wi is the weight of each indicator. When the E value is lower than the preset threshold (0.75), the system will trigger an adjustment mechanism, re-analyze the risk factors, and optimize the standardized measures. These feedback data are also used for the iterative optimization of the model training and risk assessment module, forming a complete closed-loop of 'detection - evaluation - execution - feedback'.

[0057] The feature selection sub-module selects feature variables that have a significant impact on the quality and safety risk assessment of agricultural and food products from the preprocessed data through feature selection, calculates the correlation between each feature variable and the target variable, and selects the feature variables with a significant impact.

[0058] The formula representation of feature selection is: where represents the correlation between the feature variable and the target variable, represents the feature variable, represents the target variable, and respectively represent the means of the feature variable and the target variable, is the number of samples.

[0059] In the actual application of an export-oriented fruit and vegetable enterprise, after the system was deployed and run for 6 months, the export return rate of the products decreased from 2.8% to 0.6%, and the losses caused by trade technical barriers decreased by 78%. The system gave an early warning of the changes in the revised pesticide residue limit standards in the European Union, enabling the enterprise to complete production adjustments 45 days before the formal implementation of the new standards, avoiding potential losses of approximately 2 million yuan. Through the analysis of historical data, the system identified 10 high-risk pesticide varieties and 14 key control points, and optimized the production standards accordingly. The first-pass inspection qualification rate of the products increased by 15.3 percentage points, and the inspection cost decreased by 31.7%.

[0060] The specific implementation manners described above further elaborate on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific implementation manners of the present invention and are not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A quality and safety monitoring and standardization system for agricultural and food products based on technical trade measures, characterized in that, The system includes a data acquisition module, a data processing module, a model training and risk assessment module, and a standardization execution module that are connected in sequence; The data acquisition module is used to obtain basic data from multiple data sources. The basic data includes technical trade measure data related to the quality and safety of agricultural and food products and the production control sequence data of the agricultural and food products. The technical trade measure data at least includes product standards, inspection and quarantine requirements, certification systems, and label regulations; and send the collected basic data to the data processing module; The data preprocessing module is used to clean, transform, and integrate the received basic data to form preprocessed data with a unified data format, and send it to the model training and risk assessment module; The model training and risk assessment module is used to conduct quality and safety risk assessment on agricultural and food products based on the preprocessed data. The model training and risk assessment module is also used to train a prediction model based on the preprocessed data and quantitatively predict the risk of agricultural and food products at the next moment based on the prediction model; The standardization execution module is used to formulate and execute corresponding standardization measures according to the risk assessment results.

2. The quality and safety monitoring and standardization system for agricultural and food products based on technical trade measures according to claim 1, characterized in that, The data acquisition module at least includes: a web crawler sub-module, a data interface sub-module, and a user input sub-module; The web crawler sub-module is used to crawl public technical trade measure data from the Internet; the data interface sub-module is used to interface with the data systems of production enterprises, government departments, and industry associations to obtain the production control sequence data of agricultural and food products and proprietary data related to technical trade measures; the user input sub-module is used to collect data manually input by users.

3. The quality safety monitoring and standardization system for agricultural and food products based on technical trade measures according to claim 2, characterized in that, The model training and risk assessment module includes a risk assessment model based on multivariate time series analysis. The model uses the following formula for risk assessment: Among them, represents the risk assessment value at time , is the constant term of the model, , is the regression coefficient of each variable , is the number of input variables, represents the input variables at time , including technical trade measure data and production control sequence data; is the error term at time , which follows a normal distribution; In addition, the model also includes an autoregressive moving average sub-model for capturing the autocorrelation of the data: Among them, is the coefficient of the autoregressive part; is the coefficient of the moving average part; and are the orders of autoregression and moving average respectively.

4. The quality safety monitoring and standardization system for agricultural and food products based on technical trade measures according to claim 3, characterized in that, The model training and risk assessment module includes: a data division sub-module, a feature selection sub-module, and a model evaluation sub-module; The data division sub-module is used to divide the preprocessed data into a training set and a test set for model training and verification; The feature selection sub-module is used to select feature variables that have a significant impact on the quality and safety risk assessment of agricultural and food products from the preprocessed data; The model evaluation sub-module is used to evaluate the trained model with the test set data, and calculate evaluation indicators such as the accuracy, recall rate, and F1 score of the model to determine the performance and reliability of the model; The model training and risk assessment module is trained based on the training set data using multivariate time series analysis and the autoregressive moving average sub-model, and the model is optimized and adjusted according to the evaluation indicators of the model evaluation sub-module.

5. The quality safety monitoring and standardization system for agricultural and food products based on technical trade measures according to claim 4, characterized in that, Quantitatively predicting the risk of agro-food products at the next moment based on the prediction model includes: extracting time series data from the preprocessed data and inputting the time series data into the trained prediction model; calculating the risk assessment value at the next moment ; outputting the risk assessment value , as the quantitative risk prediction result at the next moment.

6. The quality and safety monitoring and standardization system for agricultural and food products based on technical trade measures according to claim 5, characterized in that, The standardization execution module formulates standardization measures based on fuzzy logic reasoning, and the fuzzy logic reasoning is expressed as: Among them, represents the output value of the standardization measure, represents the input variable corresponding fuzzy membership function; represents the output value of each fuzzy rule; The rules used in fuzzy logic reasoning are expressed as: ; where and represent input variables; and represent the fuzzy sets of the input variables; represents the fuzzy set of the output variable.

7. The quality safety monitoring and standardization system for agricultural and food products based on technical trade measures according to claim 6, characterized in that, The standardization execution module is also used to feedback the effect of the executed standardization measures to the model training and risk assessment module. The model training and risk assessment module adjusts and optimizes the model according to the feedback information, regularly updates the data and retrains the model.

8. The quality and safety monitoring and standardization system for agricultural and food products based on technical trade measures according to claim 7, characterized in that, The feature selection sub-module selects feature variables that have a significant impact on the quality and safety risk assessment of agro-food products from the preprocessed data through feature selection, calculates the correlation between each feature variable and the target variable, and selects the feature variables with a significant impact. The formula for feature selection is expressed as: Among them, represents the correlation between the feature variable and the target variable, represents the feature variable, represents the target variable, and respectively represent the means of the feature variable and the target variable, is the number of samples.

Images