Method and device for detecting quality of finished aluminum foil meal box

Abstract

The invention discloses a quality detection method and device for an aluminum foil meal box finished product, and relates to the field of chemical analysis and test methods. The method comprises the following steps: receiving a food safety performance detection instruction; the casual inspection aluminum foil meal box is obtained; performing confidence component safety detection and confidence microorganism safety detection on the casual inspection aluminum foil meal box according to a meal box component mussel detection model to generate a meal box component safety detection result and a meal box microorganism safety detection result; carrying out mobility test and migration safety identification on the casual inspection aluminum foil meal box to generate a meal box migration safety detection result; and drawing a lunch box safety performance detection cloud picture. By adopting the method, the technical problem that in the prior art, as the food safety performance detection precision of the finished aluminum foil meal box is low, the food safety of the finished aluminum foil meal box cannot be accurately detected, and the health of a user is harmed is solved; the technical effect of accurately detecting the food safety of the finished aluminum foil meal box is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of chemical analysis and testing methods, and specifically to a quality inspection method and device for finished aluminum foil lunch box products. Background Art

[0002] As a common food packaging container in the modern catering industry, the quality and safety of aluminum foil lunch boxes are of vital importance to consumers. Therefore, quality inspection of finished aluminum foil lunch boxes is an important part of ensuring that they comply with relevant standards and regulations and protecting the rights and interests of consumers. In terms of the development of quality inspection technology, with the continuous breakthroughs in the fields of materials science, instrumental analysis and information technology, the quality inspection methods of finished aluminum foil lunch boxes have also been greatly improved. For example, advanced material analysis technology can accurately determine the chemical composition and physical properties of aluminum foil lunch boxes to evaluate their quality and safety; high-precision instrumental analysis equipment can detect harmful substances that may be released by aluminum foil lunch boxes during food contact to ensure that they meet food safety standards. Although the existing technical background provides strong support for the quality inspection of finished aluminum foil lunch boxes, there are still some challenges and needs. The existing quality inspection methods have low accuracy in detecting the food safety performance of finished aluminum foil lunch boxes, which makes it impossible to accurately detect the food safety of finished aluminum foil lunch boxes, which easily leads to technical problems that endanger the health of users.

[0003] In summary, the prior art has a technical problem that the food safety of the finished aluminum foil lunch box cannot be accurately detected due to the low accuracy of the food safety performance detection of the finished aluminum foil lunch box, resulting in harm to the health of users. Summary of the invention

[0004] Based on this, it is necessary to provide a quality inspection method and device for finished aluminum foil lunch boxes that can accurately detect the food safety of finished aluminum foil lunch boxes by introducing migration detection to address the above technical problems.

[0005] In a first aspect, a quality inspection method for finished aluminum foil lunch boxes is provided, the method comprising: receiving a food safety performance inspection instruction for a batch of finished aluminum foil lunch boxes, wherein the food safety performance inspection instruction comprises a batch full inspection instruction / a batch sampling inspection instruction, and the batch sampling inspection instruction comprises a predetermined batch sampling inspection coefficient; when the food safety performance inspection instruction is a batch sampling inspection instruction, sampling and pre-processing the batch of finished aluminum foil lunch boxes according to the predetermined batch sampling inspection coefficient to obtain the sampled aluminum foil lunch boxes, wherein the pre-processing comprises a cleaning treatment; performing a trusted component safety inspection on the sampled aluminum foil lunch boxes according to a lunch box component harm detection model to generate a lunch box component safety inspection result; performing a trusted microbial safety inspection on the sampled aluminum foil lunch boxes to generate a lunch box microbial safety inspection result; performing a migration test and migration safety identification on the sampled aluminum foil lunch boxes to generate a lunch box migration safety inspection result; and drawing a lunch box safety performance inspection cloud map according to the lunch box component safety inspection result, the lunch box microbial safety inspection result and the lunch box migration safety inspection result.

[0006] In a second aspect, a quality inspection device for finished aluminum foil lunch boxes is provided, the device comprising: a food safety performance inspection instruction receiving module, the food safety performance inspection instruction receiving module is used to receive food safety performance inspection instructions for batch aluminum foil lunch box finished products, wherein the food safety performance inspection instructions include batch full inspection instructions / batch sampling instructions, and the batch sampling instructions include a predetermined batch sampling coefficient; a sampling aluminum foil lunch box acquisition module, the sampling aluminum foil lunch box acquisition module is used to sample and pre-process the batch of aluminum foil lunch box finished products according to the predetermined batch sampling coefficient when the food safety performance inspection instruction is a batch sampling instruction, and obtain the sampling aluminum foil lunch box, wherein the pre-processing includes cleaning treatment; a trusted component safety detection module, the trusted component safety detection module is used to A lunch box ingredient hazard detection model is used to perform a trusted ingredient safety test on the sampled aluminum foil lunch boxes to generate a lunch box ingredient safety test result; a trusted microbial safety detection module is used to perform a trusted microbial safety test on the sampled aluminum foil lunch boxes to generate a lunch box microbial safety test result; a lunch box migration safety test result generation module is used to perform a migration test and migration safety identification on the sampled aluminum foil lunch boxes to generate a lunch box migration safety test result; a lunch box safety performance test cloud map drawing module is used to draw a lunch box safety performance test cloud map based on the lunch box ingredient safety test result, the lunch box microbial safety test result and the lunch box migration safety test result.

[0007] The above-mentioned quality inspection method and device for finished aluminum foil lunch box products use this method to solve the technical problem in the prior art that the food safety of the finished aluminum foil lunch box products cannot be accurately inspected due to the low accuracy of food safety performance inspection of the finished aluminum foil lunch box products, resulting in harm to the health of users. By introducing migration inspection, the technical effect of accurately inspecting the food safety of the finished aluminum foil lunch box products is achieved.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A schematic diagram of a process for quality inspection of a finished aluminum foil lunch box in one embodiment; Figure 2 A schematic diagram of a process flow of a food contact test for a quality inspection method of a finished aluminum foil lunch box in one embodiment; Figure 3 The present invention is a structural block diagram of a quality inspection device for a finished aluminum foil lunch box in one embodiment.

[0010] Explanation of the accompanying drawings: food safety performance testing instruction receiving module 11, sampling aluminum foil lunch box obtaining module 12, trusted ingredient safety testing module 13, trusted microbial safety testing module 14, lunch box migration safety testing result generating module 15, lunch box safety performance testing cloud map drawing module 16. DETAILED DESCRIPTION

[0011] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0012] like Figure 1 As shown, the present application provides a method for quality inspection of finished aluminum foil lunch boxes, the method comprising: Receiving a food safety performance test instruction for a batch of finished aluminum foil lunch boxes, wherein the food safety performance test instruction includes a batch full inspection instruction / batch sampling inspection instruction, and the batch sampling inspection instruction includes a predetermined batch sampling inspection coefficient; A finished aluminum foil lunch box is a tableware made of high-quality aluminum foil material, which is usually used in food cooking, baking, freezing, preservation and other occasions; the quality inspection method refers to the means of inspecting the finished aluminum foil lunch box, covering multiple aspects to ensure its safety, hygiene and functionality. This application provides a quality inspection method for a finished aluminum foil lunch box, conducts food safety performance inspection on the finished aluminum foil lunch box, introduces the migration test of the aluminum foil lunch box, ensures that it will not have adverse effects on food during use, improves the efficiency and accuracy of quality inspection, and provides a strong guarantee for the development of the aluminum foil lunch box industry.

[0013] Receive food safety performance test instructions for a batch of finished aluminum foil lunch boxes, where the batch of finished aluminum foil lunch boxes refers to multiple finished aluminum foil lunch boxes to be tested from the same production batch, where the food safety performance test instructions include batch full inspection instructions / batch sampling instructions, batch full inspection instructions refer to instructions for testing all the finished aluminum foil lunch boxes in the batch, batch sampling instructions refer to instructions for obtaining representative samples from the finished aluminum foil lunch boxes in the batch for testing, and the batch sampling instructions include a predetermined batch sampling coefficient, which is used to calculate the number of samples to be sampled. The number of samples to be sampled is calculated based on the predetermined batch sampling coefficient and the number of the whole batch of finished aluminum foil lunch boxes, which can ensure that the number of samples sampled is sufficiently representative and can reflect the quality status of the whole batch of products. The above method provides a basis for the subsequent processing of the batch of finished aluminum foil lunch boxes.

[0014] When the food safety performance detection instruction is a batch sampling instruction, sampling and pre-processing the finished aluminum foil lunch box batch according to the predetermined batch sampling coefficient to obtain the sampled aluminum foil lunch box, wherein the pre-processing includes cleaning; When the food safety performance test instruction is a batch sampling instruction, the batch of aluminum foil lunch box finished products is sampled according to the predetermined batch sampling coefficient, that is, according to the batch sampling coefficient and the total number of the batch of aluminum foil lunch box finished products, the specific number of sampling to be performed is calculated. The batch sampling coefficient can be determined based on past experience, the historical qualified rate of the product and the specific requirements of the customer. In the whole batch of aluminum foil lunch box finished products, a random sampling method is adopted to ensure that the probability of each aluminum foil lunch box being selected is equal, so as to ensure the representativeness and fairness of the sampling results. The batch of aluminum foil lunch box finished products is pre-processed according to the predetermined batch sampling coefficient to obtain the sampling aluminum foil lunch box, wherein the pre-processing includes cleaning treatment, that is, the aluminum foil lunch box sample is first cleaned. This includes removing impurities such as dust and oil on the surface to ensure the accuracy of subsequent test results. The sampling aluminum foil lunch box refers to the aluminum foil lunch box to be tested after sampling and pre-processing of the batch of aluminum foil lunch box finished products, which is recorded as the sampling aluminum foil lunch box. Through the above method, the accuracy and reliability of the test results are ensured. Through scientific sampling and pretreatment processes, the food safety performance of the entire batch of finished aluminum foil lunch boxes can be effectively evaluated, providing strong guarantees for product quality.

[0015] Performing a trusted component safety test on the sampled aluminum foil lunch boxes according to a lunch box component hazard detection model to generate a lunch box component safety test result; Select an established and trained lunch box ingredient harm detection model, which should be able to accurately identify harmful substances in aluminum foil lunch boxes and evaluate their potential impact on food safety. Perform a confidence ingredient safety test on the sampled aluminum foil lunch boxes according to the lunch box ingredient harm detection model, that is, analyze the sampled aluminum foil lunch boxes according to the lunch box ingredient harm detection model, be able to identify and measure the chemical components in the aluminum foil lunch boxes, and identify harmful substances in the aluminum foil lunch boxes according to the test results, and perform quantitative analysis on them. Based on the test data and model prediction results, perform a confidence assessment on the safety of the aluminum foil lunch box ingredients, and generate a lunch box ingredient safety test result, which refers to the result obtained by integrating the lunch box ingredient harm identification result and the lunch box ingredient safety test index of the sampled aluminum foil lunch boxes. Through the confidence ingredient safety test, the ingredient safety of the sampled aluminum foil lunch boxes can be comprehensively evaluated, providing a strong guarantee for product quality control and food safety.

[0016] Perform multiple component tests on the sampled aluminum foil lunch boxes to obtain multiple lunch box component test sets; Performing standardization processing and central value calculation on the multiple lunch box component detection sets to obtain a confident lunch box component detection set; Constructing the lunch box ingredient harm detection model, wherein the lunch box ingredient harm detection model comprises an input layer, a lunch box ingredient harm identification layer, a lunch box ingredient safety detection layer and an output layer; Inputting the confident lunch box ingredient detection set into the lunch box ingredient harmful identification layer, and outputting the lunch box ingredient harmful identification result; Inputting the harmfulness identification result of the lunch box ingredients into the lunch box ingredient safety detection layer to obtain a lunch box ingredient safety detection index; The lunch box ingredient hazard identification result and the lunch box ingredient safety detection index are added to the lunch box ingredient safety detection result.

[0017] According to a predetermined sampling scheme, a certain number of samples are randomly selected from the entire batch of aluminum foil lunch boxes, namely the sampled aluminum foil lunch boxes, and each sampled aluminum foil lunch box is subjected to multiple component tests to ensure the stability and reliability of the test results. This may include using different detection methods to perform tests at different time points. Each test will generate a lunch box component test set, which refers to a collection of test results for lunch box components, including volatile organic compound tests, heavy metal tests, etc. After multiple tests, multiple lunch box component test sets will be obtained. The collected multiple lunch box component test sets are cleaned to remove outliers, duplicate data or missing values, and the cleaned data are standardized to eliminate the dimensional differences between different test sets so that the data are comparable; the centralized values ​​of each component in multiple test sets, such as the mean, median, etc., are calculated to obtain more representative data. After standardization and centralized value calculation, a confident lunch box component test set is obtained, and the confident lunch box component test set will be used for subsequent model testing. A lunch box ingredient harmfulness detection model is constructed, which includes an input layer, a lunch box ingredient harmfulness identification layer, a lunch box ingredient safety detection layer and an output layer; a confident lunch box ingredient detection set is received as input data of the model, and in the lunch box ingredient harmfulness identification layer, harmful substances or ingredients present in the lunch box are identified through a machine learning algorithm; in the lunch box ingredient safety detection layer, the safety of the lunch box ingredients is evaluated based on the identified harmful substances or ingredients in combination with safety standards or thresholds, and a lunch box ingredient safety detection index is calculated. The confidence lunch box ingredient detection set is input into the lunch box ingredient harm identification layer, and the model will output the lunch box ingredient harm identification result, wherein the lunch box ingredient harm identification result refers to the evaluation result of the potential harm that these ingredients may cause to human health or the environment after the aluminum foil lunch box is tested. The evaluation result is usually based on multiple test items and standards, including chemical composition testing, heavy metal testing, and environmental protection and degradation performance testing. The lunch box ingredient harm identification result is input into the lunch box ingredient safety detection layer, and the model will calculate the lunch box ingredient safety detection index. The lunch box ingredient harm identification result and the lunch box ingredient safety detection index are added to the lunch box ingredient safety detection result to form a complete detection result report, which is recorded as the lunch box ingredient safety detection result. Through the above method, the lunch box ingredient harm detection model is constructed using the multiple ingredient detection data of the sampled aluminum foil lunch box, and the lunch box ingredient safety detection result is obtained, providing a strong guarantee for food safety.

[0018] Building a lunch box ingredient harmfulness detection model architecture, wherein the lunch box ingredient harmfulness detection model architecture includes an input layer, a first hidden layer, a second hidden layer and an output layer; Load the records of the confident lunch box ingredient detection set, the records of the harmful identification results of the lunch box ingredients, and the records of the lunch box ingredient safety detection index; The confident lunch box ingredient detection set records are used as input data, and the lunch box ingredient harmful identification result records are used as output data, and the heterogeneous machine learning channel is supervised for learning. When the harmful identification learning coefficient satisfies the harmful identification learning constraint, a lunch box ingredient harmful identifier is generated, wherein the heterogeneous machine learning channel includes a plurality of machine learning methods; Based on the heterogeneous machine learning channel, supervised learning is performed on the lunch box ingredient hazard identification result record and the lunch box ingredient safety detection index record, and when the safety detection learning coefficient satisfies the safety detection learning constraint, a lunch box ingredient safety detector is generated; The lunch box ingredient harmfulness identifier and the lunch box ingredient safety detector are used in combination with the lunch box ingredient harmfulness detection model architecture to generate the lunch box ingredient harmfulness detection model.

[0019] Build a lunch box ingredient harmful detection model architecture, wherein the lunch box ingredient harmful detection model architecture includes an input layer, a first hidden layer, a second hidden layer and an output layer. The input layer is responsible for receiving the trusted lunch box ingredient detection set records as the input data of the model. According to the number of features of the input data, the number of neurons in the input layer is designed. The first hidden layer is used to extract the features of the input data and convert it into a higher-level representation. Select the appropriate activation function and number of neurons to ensure that the model can learn the complex patterns of the data. In this application, the lunch box ingredient harmful identifier is embedded in the first hidden layer, and the second hidden layer further processes the output of the first hidden layer to extract deeper features. It is also necessary to select a suitable activation function and number of neurons. In this application, the lunch box ingredient safety detector is embedded in the second hidden layer, and the output layer is responsible for generating the final lunch box ingredient harmful identification results and lunch box ingredient safety detection index. According to the task requirements, the output layer can be designed as two independent output nodes, corresponding to the harmful identification results and the safety detection index respectively. Load the records of the confident lunch box ingredient detection set, the records of the harmful identification results of the lunch box ingredients, and the records of the safety detection index of the lunch box ingredients. The above records are the results of the detection of aluminum foil lunch boxes of the same specifications in the past. They can be obtained through queries to build a heterogeneous machine learning channel containing multiple machine learning methods. These machine learning methods can include traditional machine learning algorithms, such as support vector machines, decision trees, etc., as well as deep learning algorithms, such as neural networks, convolutional neural networks, etc. The records of the confident lunch box ingredient detection set are used as input data, and the records of the harmful identification results of the lunch box ingredients are used as output data to perform supervised learning training on the heterogeneous machine learning channel. By adjusting the parameters and structure of the model, the model can accurately identify the harmful ingredients in the lunch box. During the training process, the harmful identification learning coefficient is introduced to evaluate the learning effect of the model. When the coefficient meets the preset harmful identification learning constraints, it is considered that the model has learned enough harmful identification capabilities and a lunch box ingredient harmful identifier can be generated. Based on the heterogeneous machine learning channel, the records of the harmful identification results of the lunch box ingredients and the records of the safety detection index of the lunch box ingredients are used for further supervised learning training. The safety detection learning coefficient is also introduced to evaluate the learning effect of the model in safety detection. When the coefficient meets the preset safety detection learning constraints, it is considered that the model has sufficient safety detection capabilities and a lunch box ingredient safety detector can be generated. The generated lunch box ingredient harm identifier and lunch box ingredient safety detector are embedded in the first hidden layer and the second hidden layer, and integrated with the lunch box ingredient harm detection model architecture to obtain the lunch box ingredient harm detection model. Ensure that the various parts can work together to realize the function of lunch box ingredient harm detection. Through the above method, a lunch box ingredient harm detection model based on heterogeneous machine learning channels is built to achieve effective detection of the safety and harmfulness of lunch box ingredients.

[0020] Conducting a reliable microbial safety test on the sampled aluminum foil lunch boxes to generate a lunch box microbial safety test result; Conduct a confidence microbial safety test on the sampled aluminum foil lunch boxes, and generate the lunch box microbial safety test results. Compare the detected microbial types and quantities with the relevant food safety standards to evaluate the microbial safety of the aluminum foil lunch boxes. Consider a variety of factors, conduct a confidence analysis on the test results, and organize the results of microbial culture, counting, identification and safety assessment to form a detailed test report, i.e., the lunch box microbial safety test results. Through the above method, the sampled aluminum foil lunch boxes can be tested for confidence microbial safety, and accurate and reliable lunch box microbial safety test results can be generated, which helps to ensure the microbial safety of the aluminum foil lunch boxes.

[0021] Perform multiple microbial tests on the sampled aluminum foil lunch boxes to obtain multiple lunch box microbial test sets; Performing standardization processing and central value calculation on the multiple lunch box microorganism detection sets to obtain a confident lunch box microorganism detection set; Performing microbial damage detection on the confident lunch box microbial detection set according to the lunch box microbial damage identification sub-model to generate a lunch box microbial damage identification result; Performing a microbial safety evaluation on the lunch box microbial hazard identification result according to the lunch box microbial safety detection sub-model to generate a lunch box microbial safety evaluation coefficient; The lunch box microbial hazard identification result and the lunch box microbial safety evaluation coefficient are added to the lunch box microbial safety detection result.

[0022] The sampled aluminum foil lunch boxes are subjected to multiple microbial tests to ensure the stability and reliability of the test results. Each test will conduct a detailed analysis of the types and quantities of microorganisms in the lunch boxes; each microbial test will generate a lunch box microbial test set, including the test results of various microorganisms. Through multiple tests, multiple lunch box microbial test sets will be obtained. The collected multiple lunch box microbial test sets are cleaned, outliers or duplicate data are removed, and standardized to eliminate the dimensional differences between different test sets, and the central values ​​of each microbial species in multiple test sets are calculated to obtain more representative data. It helps to reduce the randomness of single test results and improve the reliability of the overall data. After standardization and central value calculation, a confident lunch box microbial test set is obtained. The confident lunch box microbial test set will be used as input data for subsequent microbial hazard identification and safety evaluation. A sub-model for identifying the harm caused by microorganisms in lunch boxes is constructed, and microorganism harm detection is performed on the microorganism detection set of the confident lunch box. The types of microorganisms that are harmful to human health can be identified, and the results of identification of microorganisms in lunch boxes are output. The results of identification of microorganisms in lunch boxes include the types of microorganisms with harm risks. Based on the results of identification of microorganisms in lunch boxes, the microorganism safety detection sub-model of lunch boxes is used to perform microbial safety evaluation. For example, the types and quantities of microorganisms and the thresholds of relevant safety standards are used to quantitatively evaluate the microbial safety of lunch boxes. According to the output of the safety detection sub-model, a microorganism safety evaluation coefficient of lunch boxes is generated. This coefficient will reflect the performance of lunch boxes in terms of microbial safety and can be used as an important indicator for evaluating the overall safety of lunch boxes. For example, the number and types of microorganisms are set as the thresholds of safety standards, and the types of microbial safety detection are weighted to generate a microbial safety evaluation coefficient for lunch boxes; the results of identification of microorganisms in lunch boxes and the microbial safety evaluation coefficient of lunch boxes are integrated to form a complete microbial safety detection result of lunch boxes. Through the above method, the confident microbial safety detection of sampled aluminum foil lunch boxes is realized, and comprehensive and accurate microbial safety detection results of lunch boxes are generated, which is helpful to improve the quality and safety level of aluminum foil lunch boxes.

[0023] Performing migration test and migration safety identification on the sampled aluminum foil lunch boxes, and generating lunch box migration safety test results; Migration testing of aluminum foil lunch boxes is an important step to ensure that they will not have adverse effects on food during use. This involves the detection of substances released into food after the lunch box comes into contact with food. Migration testing refers to contacting the pretreated aluminum foil lunch box sample with the test solution under set temperature, time and other conditions to simulate the migration situation during actual use. After the contact is completed, the substances that may migrate into the lunch box are extracted from the test solution, and the extracted migrants are qualitatively and quantitatively analyzed using appropriate analytical methods to determine their types and contents. Migration safety identification refers to comparing the detection results of migrants with relevant food safety standards to evaluate the safety of migrants. Combined with the nature of the migrants, conditions of use and other factors, a risk assessment of the migration safety of aluminum foil lunch boxes is conducted to determine whether there are potential health risks. The migration test data, qualitative and quantitative analysis results of migrants, safety standard comparison results and risk assessment results are collated to form a detailed test report, namely the lunch box migration safety test results. Through the above method, the sampled aluminum foil lunch boxes can be subjected to migration test and migration safety identification, and accurate and reliable lunch box migration safety test results can be generated, which helps to ensure that the aluminum foil lunch boxes will not pose a potential threat to human health during use.

[0024] According to the multidimensional migration test scenario indicators, A migration test scenarios are constructed, wherein the multidimensional migration test scenario indicators include simulated food, simulated contact time, simulated contact temperature, simulated contact area and simulated contact environment, and A is a positive integer greater than 1; Extracting the ath migration test scenario according to the A migration test scenarios, and reading the detection information of the simulated food in the ath migration test scenario to obtain the ath pre-food detection data, wherein a belongs to A and a is a positive integer; Perform the food contact test of the sampled aluminum foil lunch box according to the a migration test scenario to generate a post-food test data; Performing substance migration detection on the ath post-food detection data based on the ath pre-food detection data to generate the ath food migration detection data; Performing a substance migration safety evaluation based on the a-th food migration detection data to generate a a-th food migration safety evaluation coefficient; The ath food migration detection data and the ath food migration safety evaluation coefficient are added to the lunch box migration safety detection result.

[0025] Determine the multidimensional migration test scenario indicators, including simulated food, simulated contact time, simulated contact temperature, simulated contact area and simulated contact environment, wherein the simulated food includes the food that the aluminum foil lunch box contacts during actual use, and the multidimensional migration test scenario indicators will be used to simulate various conditions of the lunch box during actual use. According to the determined indicators, build A different migration test scenarios. Each scenario will simulate different foods, contact time, temperature, area and environmental conditions. Select the ath scenario from the A migration test scenarios as the current test object, read the detection information of the simulated food in the ath migration test scenario, and this information will be used as the pre-food detection data for comparison with the subsequent test data. Select a representative aluminum foil lunch box sample and prepare for food contact testing. Under the conditions of the ath migration test scenario, the aluminum foil lunch box sample is contacted with the simulated food to simulate the actual use process. After the contact is completed, the simulated food is tested to generate post-food detection data. The post-food detection data will reflect the substance migration that may occur after the aluminum foil lunch box contacts the food. Based on the a-th pre-food test data and the a-th post-food test data, a material migration test is performed, and the material migration test identifies the types and quantities of substances that may migrate by comparing the data before and after. According to the results of the migration test, combined with relevant food safety standards and regulations, a material migration safety evaluation is performed. The a-th food migration safety evaluation coefficient is generated by calculation or scoring. The a-th food migration safety evaluation coefficient will reflect the migration safety level of the aluminum foil lunch box in this test scenario. The a-th food migration test data and the a-th food migration safety evaluation coefficient are added to the lunch box migration safety test results, and the integrated lunch box migration safety test results are output, so that the migration safety performance of the aluminum foil lunch box in different test scenarios can be understood, and a basis for subsequent decisions and measures can be provided. According to the above method, a comprehensive migration test and migration safety identification are performed on the sampled aluminum foil lunch boxes according to the multi-dimensional migration test scenario indicators. This helps to ensure the safety of the lunch boxes during actual use.

[0026] like Figure 2 As shown, after completing the food contact test of the a-1 migration test scenario, the image data of the sampled aluminum foil lunch box is collected in real time to obtain the monitoring data of the lunch box after the a-1 test; Performing a lunch box cleaning demand analysis based on the lunch box monitoring data after the a-1 test to generate a lunch box cleaning demand coefficient; Determining whether the lunch box cleaning requirement coefficient is less than the lunch box cleaning requirement threshold; If the lunch box cleaning requirement coefficient is greater than / equal to the lunch box cleaning requirement threshold, a lunch box cleaning instruction is generated; Clean the sampled aluminum foil lunch boxes according to the lunch box cleaning instructions, and conduct food contact tests on the sampled aluminum foil lunch boxes after the cleaning process according to the a-th migration test scenario.

[0027] After completing the food contact test of the a-1-th migration test scenario, immediately use an image acquisition device, such as a high-definition camera, to perform real-time image acquisition on the sampled aluminum foil lunch boxes, ensuring that the acquired images are clear and comprehensive, capable of reflecting the internal and external surface states of the lunch boxes. Import the acquired image data into a computer, and preprocess the images through image processing software, such as denoising and enhancing contrast, in order to better extract the characteristic information of the lunch boxes. After image processing, extract key information such as stains and residues on the surface of the lunch boxes to form the lunch box monitoring data after the a-1-th test, which will be used for subsequent analysis of the lunch box cleaning requirements. According to the lunch box monitoring data after the a-1-th test, analyze the stain degree, residue types and quantities on the surface of the lunch boxes, evaluate the cleaning status of the lunch boxes. Based on the analysis results, combined with the preset cleaning requirement evaluation criteria, calculate the lunch box cleaning requirement coefficient, and the lunch box cleaning requirement coefficient will reflect the degree of demand for the cleaning process in the current state of the lunch box. Compare the calculated lunch box cleaning requirement coefficient with the preset lunch box cleaning requirement threshold, and the lunch box cleaning requirement threshold is usually determined according to factors such as food safety standards and lunch box usage scenarios. If the lunch box cleaning requirement coefficient is greater than or equal to the cleaning requirement threshold, it indicates that the lunch box needs to be cleaned. At this time, generate a lunch box cleaning instruction, and clean the sampled aluminum foil lunch boxes according to the lunch box cleaning instruction, using appropriate cleaning equipment and methods to clean the sampled aluminum foil lunch boxes. Ensure that no new pollution sources are introduced during the cleaning process, and the cleaning effect meets food safety requirements. According to the requirements of the a-th migration test scenario, conduct food contact tests on the sampled aluminum foil lunch boxes after the cleaning process. Ensure that the test conditions are consistent with the previous test for accurate comparative analysis. Through the above method, the cleaning status of the sampled aluminum foil lunch boxes is monitored in real time, and timely cleaning is carried out according to the cleaning requirements. This helps to ensure that the lunch boxes are in good cleaning condition before each test, thereby improving the accuracy of migration tests and migration safety identification. At the same time, it also provides safety guarantees for the subsequent use of the lunch boxes.

[0028] Conduct a substance migration comparison between the a-th post-food test data based on the a-th pre-food test data to generate the a-th food migration comparison result; Conduct lunch box correlation analysis based on the a-th food migration comparison result to generate the a-th correlated migration comparison result that meets the correlation constraints; Read the lunch box migration data of the sampled aluminum foil lunch boxes according to the a-1-th migration test scenario to generate the a-1-th lunch box migration data; Perform migration impact identification on the a-th associated migration comparison result according to the a-1th lunch box migration data to generate a lunch box migration impact identification result; The ath associated migration comparison result is corrected according to the lunch box migration impact identification result to obtain the ath food migration detection data.

[0029] Using the a-th pre-food test data as a reference, the a-th post-food test data is compared for material migration. In order to identify which substances migrated from the aluminum foil lunch box to the food during the food contact test, the comparison results will generate the a-th food migration comparison results, which lists the types and quantities of migrating substances in detail; the a-th food migration comparison results are analyzed for lunch box association. The purpose of this step is to find out the association between the migrating substances and the characteristics of the aluminum foil lunch box, such as material, production process, etc. Through association analysis, the migrating substances that are closely related to the characteristics of the lunch box and meet the association constraints can be screened out to generate the a-th association migration comparison results. Before the migration impact identification, the lunch box migration data of the a-1 migration test scenario is read, including the migration performance of the aluminum foil lunch box in the previous test scenario, which is crucial for analyzing the impact of the migrating substances in the current test scenario. Based on the a-1 lunch box migration data, the a-th association migration comparison results are identified for migration impact. This step is to analyze the impact of the migration situation in the previous test scenario on the migrating substances in the current test scenario, such as whether the same migrating substances appear, the trend of the migration amount, etc. Through migration impact identification, the lunch box migration impact identification result is generated. According to the lunch box migration impact identification result, the a-th associated migration comparison result is corrected. In the present application, the correction includes adjusting the estimation of the migration amount, considering factors such as the interaction between different foods and lunch box materials, and adjusting the a-th associated migration comparison result to obtain the a-th food migration detection data with higher accuracy. This step is to more accurately reflect the migration performance of the aluminum foil lunch box in the current test scenario. Through data correction, more accurate and reliable a-th food migration detection data can be obtained. Through the above method, a comprehensive migration comparison, association analysis and impact identification can be performed on the sampled aluminum foil lunch boxes, so as to obtain more accurate and comprehensive migration detection data.

[0030] A lunch box safety performance test cloud chart is drawn based on the lunch box ingredient safety test results, the lunch box microbial safety test results and the lunch box migration safety test results.

[0031] It is drawn based on the results of the safety detection of the lunch box ingredients, the safety detection of the lunch box microorganisms, and the safety detection of the lunch box migration. First, since the units and dimensions of different safety detection results may be different, it is necessary to standardize the data so as to compare and display them on the same scale. Secondly, according to safety standards or regulations, classify and rate the results of each safety detection. For example, it can be divided into levels such as excellent, good, qualified, and unqualified. Then use professional data visualization tools. Take the safety of lunch box ingredients, the safety of microorganisms, and the migration safety as the three coordinate axes of the cloud map, which respectively represent the three key safety dimensions of the lunch box. According to the processed data, determine the position of each lunch box sample in the cloud map. The closer the position is to the origin of the coordinate axis, the worse the sample performs in the corresponding safety dimension; on the contrary, the better it performs. Different colors and sizes can be used to distinguish safety detection results at different levels. For example, color gradients can be used to represent the safety levels from excellent to unqualified, and the size can reflect the severity of the detection results. Add necessary labels and explanations to the cloud map to clearly explain the safety dimension represented by each coordinate axis and the meaning of the data points. Through the above method, quickly understand the performance of each lunch box sample in terms of ingredient safety, microorganism safety, and migration safety by observing the distribution of data points in the cloud map. At the same time, the overall safety performance of different samples can also be evaluated by comparing the position differences between different samples. This helps producers and consumers more intuitively understand the safety performance of the lunch box and provides a reference basis for improving the production process and selecting safe and reliable lunch box products.

[0032] When the food safety performance detection instruction is the batch full inspection instruction, traverse the food safety performance detection of the batch of aluminum foil lunch box finished products according to the batch full inspection instruction, and generate a full inspection lunch box safety performance detection cloud map.

[0033] When the food safety performance detection instruction is the batch full inspection instruction, it means that a comprehensive food safety performance detection needs to be carried out on the entire batch of aluminum foil lunch box finished products. The food safety performance detection can be carried out according to the above method. To ensure that the safety performance of all products within the batch meets relevant standards and requirements, thus ensuring the food safety of consumers. Sort, classify, and standardize all the data obtained from the traversal detection for visual display. Use data visualization tools or programming languages to draw the processed data into a full inspection lunch box safety performance detection cloud map. The full inspection lunch box safety performance detection cloud map should be able to clearly display the performance of each sample in the three dimensions of ingredient safety, microorganism safety, and migration safety. By traversing the food safety performance detection and generating a full inspection lunch box safety performance detection cloud map, the safety performance status of the batch of aluminum foil lunch box finished products can be comprehensively understood, providing an important reference basis for producers and consumers.

[0034] Such as Figure 3As shown, the embodiment of the present application includes a quality inspection device for a finished aluminum foil lunch box, the device comprising: A food safety performance detection instruction receiving module 11, wherein the food safety performance detection instruction receiving module 11 is used to receive a food safety performance detection instruction of a batch of finished aluminum foil lunch boxes, wherein the food safety performance detection instruction includes a batch full inspection instruction / batch sampling inspection instruction, and the batch sampling inspection instruction includes a predetermined batch sampling inspection coefficient; The sampling aluminum foil lunch box obtaining module 12 is used for sampling and pre-processing the batch of aluminum foil lunch box finished products according to the predetermined batch sampling coefficient to obtain the sampling aluminum foil lunch box when the food safety performance detection instruction is a batch sampling instruction, wherein the pre-processing includes cleaning; A trusted ingredient safety detection module 13, which is used to perform a trusted ingredient safety detection on the sampled aluminum foil lunch box according to a lunch box ingredient harm detection model, and generate a lunch box ingredient safety detection result; A trusted microbial safety detection module 14, which is used to perform a trusted microbial safety detection on the sampled aluminum foil lunch boxes and generate a lunch box microbial safety detection result; A lunch box migration safety test result generation module 15, the lunch box migration safety test result generation module 15 is used to perform migration test and migration safety identification on the sampled aluminum foil lunch boxes, and generate lunch box migration safety test results; The lunch box safety performance detection cloud map drawing module 16 is used to draw a lunch box safety performance detection cloud map according to the lunch box ingredient safety detection results, the lunch box microbial safety detection results and the lunch box migration safety detection results.

[0035] Furthermore, the embodiment of the present application also includes: A component detection module, the component detection module is used to perform multiple component detections on the sampled aluminum foil lunch boxes to obtain multiple lunch box component detection sets; A lunch box component detection set standardization processing module, the lunch box component detection set standardization processing module is used to perform standardization processing and central value calculation according to the multiple lunch box component detection sets to obtain a confident lunch box component detection set; A lunch box ingredient harm detection model construction module, the lunch box ingredient harm detection model construction module is used to construct the lunch box ingredient harm detection model, wherein the lunch box ingredient harm detection model includes an input layer, a lunch box ingredient harm identification layer, a lunch box ingredient safety detection layer and an output layer; a meal box ingredient harmful identification result output module, the meal box ingredient harmful identification result output module is used to input the confident meal box ingredient detection set into the meal box ingredient harmful identification layer, and output the meal box ingredient harmful identification result; A meal box ingredient harmfulness identification result input module, the meal box ingredient harmfulness identification result input module is used to input the meal box ingredient harmfulness identification result into the meal box ingredient safety detection layer to obtain a meal box ingredient safety detection index; A lunch box ingredient adding module is used to add the lunch box ingredient harmful identification result and the lunch box ingredient safety detection index to the lunch box ingredient safety detection result.

[0036] Furthermore, the embodiment of the present application also includes: A model architecture building module, wherein the model architecture building module is used to build a lunch box ingredient harm detection model architecture, wherein the lunch box ingredient harm detection model architecture includes an input layer, a first hidden layer, a second hidden layer and an output layer; A record loading module, the record loading module is used to load a confident lunch box ingredient detection set record, a lunch box ingredient harmful identification result record and a lunch box ingredient safety detection index record; A heterogeneous machine learning channel supervised learning module, wherein the heterogeneous machine learning channel supervised learning module is used to use the confident lunch box ingredient detection set records as input data and the lunch box ingredient harmful identification result records as output data to perform supervised learning on the heterogeneous machine learning channel, and when the harmful identification learning coefficient satisfies the harmful identification learning constraint, a lunch box ingredient harmful identifier is generated, wherein the heterogeneous machine learning channel includes multiple machine learning methods; A lunch box ingredient safety detector generation module, the lunch box ingredient safety detector generation module is used to perform supervised learning on the lunch box ingredient hazard identification result record and the lunch box ingredient safety detection index record based on the heterogeneous machine learning channel, and generate a lunch box ingredient safety detector when the safety detection learning coefficient meets the safety detection learning constraint; A lunch box ingredient harmfulness detection model generation module is used to generate the lunch box ingredient harmfulness detection model based on the lunch box ingredient harmfulness identifier and the lunch box ingredient safety detector in combination with the lunch box ingredient harmfulness detection model architecture.

[0037] Furthermore, the embodiment of the present application also includes: A microbial detection module, the microbial detection module is used to perform multiple microbial detections on the sampled aluminum foil lunch boxes to obtain multiple lunch box microbial detection sets; A module for obtaining a trusted lunch box microorganism detection set, the module being used to perform standardization processing and central value calculation on the multiple lunch box microorganism detection sets to obtain a trusted lunch box microorganism detection set; A module for generating a result of identifying microbial damage to a lunch box, wherein the module is used to perform a microbial damage detection on the trusted lunch box microbial detection set according to a lunch box microbial damage identification sub-model, and generate a lunch box microbial damage identification result; A lunch box microbial safety evaluation coefficient generation module, the lunch box microbial safety evaluation coefficient generation module is used to perform a microbial safety evaluation on the lunch box microbial harm identification result according to the lunch box microbial safety detection sub-model, and generate a lunch box microbial safety evaluation coefficient; A lunch box microbial safety evaluation coefficient adding module is used to add the lunch box microbial harm identification result and the lunch box microbial safety evaluation coefficient to the lunch box microbial safety detection result.

[0038] Furthermore, the embodiment of the present application also includes: A migration test scenario building module, wherein the migration test scenario building module is used to build A migration test scenarios according to multidimensional migration test scenario indicators, wherein the multidimensional migration test scenario indicators include simulated food, simulated contact time, simulated contact temperature, simulated contact area and simulated contact environment, and A is a positive integer greater than 1; A detection information reading module, the detection information reading module is used to extract the ath migration test scenario according to the A migration test scenarios, and read the detection information of the simulated food in the ath migration test scenario to obtain the ath pre-food detection data, wherein a belongs to A and a is a positive integer; A food contact test execution module, the food contact test execution module is used to perform the food contact test of the sampled aluminum foil lunch box according to the ath migration test scenario, and generate ath post-food detection data; a material migration detection module, the material migration detection module being used to perform material migration detection on the ath post-food detection data based on the ath pre-food detection data to generate the ath food migration detection data; A migration safety evaluation coefficient generation module, wherein the migration safety evaluation coefficient generation module is used to perform a substance migration safety evaluation based on the a-th food migration detection data and generate a a-th food migration safety evaluation coefficient; A food migration safety evaluation coefficient adding test result module, wherein the food migration safety evaluation coefficient adding test result module is used to add the ath food migration test data and the ath food migration safety evaluation coefficient to the lunch box migration safety test result.

[0039] Furthermore, the embodiment of the present application also includes: An image data acquisition module, wherein the image data acquisition module is used to acquire image data of the sampled aluminum foil lunch box in real time after completing the food contact test of the a-1 migration test scenario, and obtain monitoring data of the lunch box after the a-1 test; A meal box cleaning requirement coefficient generating module, the meal box cleaning requirement coefficient generating module is used to perform a meal box cleaning requirement analysis based on the meal box monitoring data after the a-1th test, and generate a meal box cleaning requirement coefficient; A cleaning requirement coefficient judgment module, the cleaning requirement coefficient judgment module is used to judge whether the lunch box cleaning requirement coefficient is less than the lunch box cleaning requirement threshold; A lunch box cleaning instruction generating module, the lunch box cleaning instruction generating module is used to generate a lunch box cleaning instruction if the lunch box cleaning requirement coefficient is greater than / equal to the lunch box cleaning requirement threshold; A food contact testing module, wherein the food contact testing module is used to clean the sampled aluminum foil lunch box according to the lunch box cleaning instruction, and to perform a food contact test on the sampled aluminum foil lunch box after the cleaning process according to the ath migration test scenario.

[0040] Furthermore, the embodiment of the present application also includes: A food migration comparison result generating module, the food migration comparison result generating module is used to perform a substance migration comparison on the ath post-food detection data according to the ath pre-food detection data, and generate an ath food migration comparison result; A meal box association analysis module, the meal box association analysis module is used to perform a meal box association analysis based on the a-th food migration comparison result, and generate a a-th association migration comparison result that satisfies the association constraint; A lunch box migration data reading module, the lunch box migration data reading module is used to read the lunch box migration data of the sampled aluminum foil lunch box according to the a-1 migration test scenario, and generate the a-1 lunch box migration data; A migration impact identification module, the migration impact identification module is used to identify the migration impact of the a-th associated migration comparison result according to the a-1th lunch box migration data, and generate a lunch box migration impact identification result; A comparison result correction module is used to correct the a-th associated migration comparison result according to the lunch box migration impact identification result to obtain the a-th food migration detection data.

[0041] Furthermore, the embodiment of the present application also includes: A food safety performance testing module, wherein when the food safety performance testing instruction is the batch full inspection instruction, the food safety performance testing module is used to perform food safety performance testing on the batch of finished aluminum foil lunch boxes according to the batch full inspection instruction to generate a cloud map of safety performance testing of full inspection lunch boxes.

[0042] Each module in the above-mentioned quality inspection device for finished aluminum foil lunch boxes can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0043] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0044] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application.

Claims

1. A quality inspection method for finished aluminum foil lunch box products, characterized in that: The method comprises: Receiving a food safety performance test instruction for a batch of finished aluminum foil lunch boxes, wherein the food safety performance test instruction includes a batch full inspection instruction / batch sampling inspection instruction, and the batch sampling inspection instruction includes a predetermined batch sampling inspection coefficient; When the food safety performance detection instruction is a batch sampling instruction, sampling and pre-processing the finished aluminum foil lunch box batch according to the predetermined batch sampling coefficient to obtain the sampled aluminum foil lunch box, wherein the pre-processing includes cleaning; Performing a trusted component safety test on the sampled aluminum foil lunch boxes according to a lunch box component hazard detection model to generate a lunch box component safety test result; Conducting a reliable microbial safety test on the sampled aluminum foil lunch boxes to generate a lunch box microbial safety test result; Performing migration test and migration safety identification on the sampled aluminum foil lunch boxes, and generating lunch box migration safety test results; A lunch box safety performance test cloud chart is drawn based on the lunch box ingredient safety test results, the lunch box microbial safety test results and the lunch box migration safety test results.

2. The method according to claim 1, characterized in that The sampled aluminum foil lunch boxes are subjected to a trusted component safety test according to the lunch box component harm detection model, and a lunch box component safety test result is generated, including: Perform multiple component tests on the sampled aluminum foil lunch boxes to obtain multiple lunch box component test sets; Performing standardization processing and central value calculation on the multiple lunch box component detection sets to obtain a confident lunch box component detection set; Constructing the lunch box ingredient harm detection model, wherein the lunch box ingredient harm detection model comprises an input layer, a lunch box ingredient harm identification layer, a lunch box ingredient safety detection layer and an output layer; Inputting the confident lunch box ingredient detection set into the lunch box ingredient harmful identification layer, and outputting the lunch box ingredient harmful identification result; Inputting the harmful identification result of the lunch box ingredients into the lunch box ingredient safety detection layer to obtain a lunch box ingredient safety detection index; The lunch box ingredient hazard identification result and the lunch box ingredient safety detection index are added to the lunch box ingredient safety detection result.

3. The method according to claim 2, characterized in that Constructing the lunch box ingredient harmfulness detection model includes: Building a lunch box ingredient harmfulness detection model architecture, wherein the lunch box ingredient harmfulness detection model architecture includes an input layer, a first hidden layer, a second hidden layer and an output layer; Load the records of the confident lunch box ingredient detection set, the records of the harmful identification results of the lunch box ingredients, and the records of the lunch box ingredient safety detection index; The confident lunch box ingredient detection set records are used as input data, and the lunch box ingredient harmful identification result records are used as output data, and the heterogeneous machine learning channel is supervised for learning. When the harmful identification learning coefficient satisfies the harmful identification learning constraint, a lunch box ingredient harmful identifier is generated, wherein the heterogeneous machine learning channel includes a plurality of machine learning methods; Based on the heterogeneous machine learning channel, supervised learning is performed on the lunch box ingredient hazard identification result record and the lunch box ingredient safety detection index record, and when the safety detection learning coefficient satisfies the safety detection learning constraint, a lunch box ingredient safety detector is generated; The lunch box ingredient harmfulness identifier and the lunch box ingredient safety detector are used in combination with the lunch box ingredient harmfulness detection model architecture to generate the lunch box ingredient harmfulness detection model.

4. The method according to claim 1, characterized in that Conduct a reliable microbial safety test on the sampled aluminum foil lunch boxes to generate lunch box microbial safety test results, including: Perform multiple microbial tests on the sampled aluminum foil lunch boxes to obtain multiple lunch box microbial test sets; Performing standardization processing and central value calculation on the multiple lunch box microorganism detection sets to obtain a confident lunch box microorganism detection set; Performing microbial damage detection on the confident lunch box microbial detection set according to the lunch box microbial damage identification sub-model to generate a lunch box microbial damage identification result; Performing a microbial safety evaluation on the lunch box microbial hazard identification result according to the lunch box microbial safety detection sub-model to generate a lunch box microbial safety evaluation coefficient; The lunch box microbial hazard identification result and the lunch box microbial safety evaluation coefficient are added to the lunch box microbial safety detection result.

5. The method according to claim 1, characterized in that The sampled aluminum foil lunch boxes are subjected to migration test and migration safety identification, and the lunch box migration safety test results are generated, including: According to the multidimensional migration test scenario indicators, A migration test scenarios are constructed, wherein the multidimensional migration test scenario indicators include simulated food, simulated contact time, simulated contact temperature, simulated contact area and simulated contact environment, and A is a positive integer greater than 1; Extracting the ath migration test scenario according to the A migration test scenarios, and reading the detection information of the simulated food in the ath migration test scenario to obtain the ath pre-food detection data, wherein a belongs to A and a is a positive integer; Perform the food contact test of the sampled aluminum foil lunch box according to the a migration test scenario to generate a post-food test data; Performing substance migration detection on the ath post-food detection data based on the ath pre-food detection data to generate the ath food migration detection data; Performing substance migration safety evaluation based on the a-th food migration detection data to generate the a-th food migration safety evaluation coefficient; The ath food migration detection data and the ath food migration safety evaluation coefficient are added to the lunch box migration safety detection result.

6. The method according to claim 5, characterized in that The food contact test of the sampled aluminum foil lunch box is performed according to the migration test scenario a, including: After completing the food contact test of the a-1 migration test scenario, the image data of the sampled aluminum foil lunch box is collected in real time to obtain the monitoring data of the lunch box after the a-1 test; Performing a lunch box cleaning demand analysis based on the lunch box monitoring data after the a-1 test to generate a lunch box cleaning demand coefficient; Determining whether the lunch box cleaning requirement coefficient is less than the lunch box cleaning requirement threshold; If the lunch box cleaning requirement coefficient is greater than / equal to the lunch box cleaning requirement threshold, a lunch box cleaning instruction is generated; The sampled aluminum foil lunch box is cleaned according to the lunch box cleaning instruction, and a food contact test is performed on the sampled aluminum foil lunch box after the cleaning according to the ath migration test scenario.

7. The method according to claim 6, characterized in that Performing substance migration detection on the ath post-food detection data based on the ath pre-food detection data to generate the ath food migration detection data includes: Performing a substance migration comparison on the ath post-food detection data according to the ath pre-food detection data to generate an ath food migration comparison result; Performing a lunch box association analysis based on the a-th food migration comparison result to generate a a-th association migration comparison result that satisfies the association constraint; Read the lunch box migration data of the sampled aluminum foil lunch box according to the a-1 migration test scenario to generate a-1 lunch box migration data; Perform migration impact identification on the a-th associated migration comparison result according to the a-1th lunch box migration data to generate a lunch box migration impact identification result; The ath associated migration comparison result is corrected according to the lunch box migration impact identification result to obtain the ath food migration detection data.

8. The method according to claim 1, characterized in that When the food safety performance test instruction is the batch full inspection instruction, the batch of finished aluminum foil lunch boxes are subjected to a food safety performance test according to the batch full inspection instruction to generate a full inspection lunch box safety performance test cloud map.

9. A quality inspection device for finished aluminum foil lunch boxes, characterized in that: The device comprises: A food safety performance detection instruction receiving module, the food safety performance detection instruction receiving module is used to receive a food safety performance detection instruction of a batch of finished aluminum foil lunch boxes, wherein the food safety performance detection instruction includes a batch full inspection instruction / batch sampling inspection instruction, and the batch sampling inspection instruction includes a predetermined batch sampling inspection coefficient; A sampling aluminum foil lunch box acquisition module, wherein the sampling aluminum foil lunch box acquisition module is used to sample and pre-process the batch of aluminum foil lunch box products according to the predetermined batch sampling coefficient to obtain the sampling aluminum foil lunch box when the food safety performance detection instruction is a batch sampling instruction, wherein the pre-processing includes cleaning; A trusted ingredient safety detection module, which is used to perform a trusted ingredient safety detection on the sampled aluminum foil lunch box according to a lunch box ingredient harm detection model, and generate a lunch box ingredient safety detection result; A trusted microbial safety detection module, which is used to perform a trusted microbial safety detection on the sampled aluminum foil lunch box and generate a lunch box microbial safety detection result; A lunch box migration safety test result generation module, the lunch box migration safety test result generation module is used to perform migration test and migration safety identification on the sampled aluminum foil lunch boxes, and generate a lunch box migration safety test result; A lunch box safety performance detection cloud map drawing module is used to draw a lunch box safety performance detection cloud map based on the lunch box ingredient safety detection results, the lunch box microbial safety detection results and the lunch box migration safety detection results.

Images