Pesticide residue detection method and system for table vinegar production

Abstract

The invention relates to the technical field of pesticide residue detection, in particular to a pesticide residue detection method and system for table vinegar production, and the method comprises the steps: obtaining a table vinegar raw material data set, carrying out adaptive clustering, and determining the sample size according to a preset proportion; the first detection end performs random sampling according to the sample size, generates a raw material danger identifier, uploads the raw material danger identifier to the first server, and predicts the vinegar production risk in combination with the monitoring data; and configuring sampling parameters, and controlling the second detection end to carry out vinegar detection. The technical problems that vinegar solution components interfere with pesticide residue detection and the pesticide residue detection accuracy is low are solved, monitoring from raw material purchasing as a starting point is achieved, a perfect raw material source tracing system is established through an informatization means, it is ensured that the source controls the pesticide residue risk, the pesticide residue detection accuracy of vinegar production is ensured, and the pesticide residue detection efficiency is improved. And an online real-time monitoring technology is adopted to capture and analyze pesticide residue information in raw materials or a processing process in real time, so that the technical effect of improving the detection efficiency is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field related to pesticide residue detection, and in particular to a method and system for detecting pesticide residues in vinegar production. Background Art

[0002] Vinegar is a widely used condiment, and its production process may be affected by pesticide residues in raw crops. Since pesticides are widely used in agricultural production, if they are not fully degraded or cleaned after harvest, vinegar products may contain a certain amount of pesticide residues. Commonly, pesticide residue detection in vinegar production usually needs to be carried out in the laboratory, including sample pretreatment, extraction, concentration, purification and other steps. After production is completed, the processed vinegar is tested for pesticide residues, but this method is often time-consuming and costly. The components of the vinegar solution interfere with the detection of pesticide residues and cannot meet the needs of real-time monitoring and rapid response.

[0003] In summary, the prior art has the technical problem that the components of the vinegar solution interfere with the detection of pesticide residues, resulting in low accuracy of the detection of pesticide residues. Summary of the invention

[0004] The present application provides a method and system for detecting pesticide residues in vinegar production, aiming to solve the technical problems in the prior art that the components of vinegar solution interfere with the detection of pesticide residues and the accuracy of pesticide residue detection is low.

[0005] In view of the above problems, the technical solution to implement this application is:

[0006] In one aspect, the present application provides a method for detecting pesticide residues in vinegar production, wherein the method is applied to a pesticide residue detection system, the pesticide residue detection system comprising a processing end, digitally communicating with a first service end, a first detection end, and a second detection end, the method comprising:

[0007] K1: Establishing a production raw material data set of vinegar, wherein the production raw material data set is basic data of vinegar raw materials obtained from the first server end after the pesticide residue detection system establishes digital communication with the first server end, and the production raw material data set includes the source location of the raw materials and the source time of the raw materials. The processing end uses the production raw material data set to perform adaptive similarity clustering of the raw materials, establish a similar cluster set, and set the sample size of the similar cluster set by a preset sampling ratio;

[0008] K2: using the sample size as a collection constraint, controlling the first detection end to perform random sampling of similar cluster sets, and generating a raw material hazard identification according to the random sampling result, and uploading the raw material hazard identification to the first service end, which is then sent to the pesticide residue detection system by the first service end, obtaining monitoring data of raw material synthesis records through the processing end, and predicting vinegar production risks using the monitoring data and the raw material hazard identification to obtain prediction results;

[0009] K3: configuring a preset ratio of random samples and weighted samples, and setting a total sampling amount, and configuring sampling parameters through the processing end according to the preset ratio, the total sampling amount and the prediction result;

[0010] K4: Controlling the second detection end to perform vinegar production detection through the sampling parameters, and generating a pesticide residue detection result, and sending the pesticide residue detection result to the first service end for storage.

[0011] On the other hand, the present application provides a pesticide residue detection system for vinegar production, wherein the pesticide residue detection system includes a processing end, digitally communicating with a first service end, a first detection end, and a second detection end, and the system includes:

[0012] A data set establishment module is used to establish a production raw material data set of vinegar, wherein the production raw material data set is basic data of vinegar raw materials obtained from the first server end after the pesticide residue detection system establishes digital communication with the first server end, and the production raw material data set includes the source location of the raw materials and the source time of the raw materials. The processing end uses the production raw material data set to perform adaptive similarity clustering of the raw materials, establish a similar cluster set, and set the sample size of the similar cluster set by a preset sampling ratio;

[0013] A random sampling module, used to control the first detection end to perform random sampling of similar cluster sets with the sample size as a collection constraint, and to generate a raw material hazard identification according to the random sampling result, and to upload the raw material hazard identification to the first service end, which is then sent to the pesticide residue detection system by the first service end, and to obtain monitoring data of raw material synthesis records through the processing end, and to perform vinegar production risk prediction based on the monitoring data and the raw material hazard identification to obtain a prediction result;

[0014] A preset ratio configuration module, used to configure a preset ratio of random samples and weighted samples, and set the total sampling amount, and configure sampling parameters according to the preset ratio, the total sampling amount and the prediction result through the processing end;

[0015] The vinegar production detection module is used to control the second detection end to perform vinegar production detection through the sampling parameters, generate pesticide residue detection results, and send the pesticide residue detection results to the first service end for storage.

[0016] In summary, one or more technical solutions provided in this application realize monitoring from the purchase of raw materials as the starting point, establish a complete raw material source traceability system through information technology, ensure the risk of pesticide residues at the source, ensure the accuracy of pesticide residue detection in vinegar production, and use online real-time monitoring technology to capture and analyze pesticide residue information in raw materials or processing in real time, thereby improving the technical effect of detection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A possible flow chart of a method for detecting pesticide residues in vinegar production is provided for this application;

[0018] Figure 2 This application provides a possible flow chart for generating raw material hazard identification in a method for detecting pesticide residues in vinegar production;

[0019] Figure 3 A possible structural schematic diagram of a pesticide residue detection system for vinegar production is provided for this application.

[0020] Description of reference numerals: data set establishment module M100, random sampling module M200, preset ratio configuration module M300, vinegar production detection module M400. DETAILED DESCRIPTION

[0021] Embodiment 1

[0022] The present application is described in detail below with reference to the accompanying drawings. Figure 1 As shown, the present application provides a method for detecting pesticide residues in vinegar production, wherein the method is applied to a pesticide residue detection system, the pesticide residue detection system includes a processing end, digitally communicating with a first service end, a first detection end, and a second detection end, and the method includes:

[0023] K1: Establishing a production raw material data set of vinegar, wherein the production raw material data set is basic data of vinegar raw materials obtained from the first server end after the pesticide residue detection system establishes digital communication with the first server end, and the production raw material data set includes the source location of the raw materials and the source time of the raw materials. The processing end uses the production raw material data set to perform adaptive similarity clustering of the raw materials, establish a similar cluster set, and set the sample size of the similar cluster set by a preset sampling ratio;

[0024] Since vinegar itself contains a variety of organic acids and compounds, it may interfere with the extraction and detection of pesticide residues. For example, the acidic substances in vinegar may corrode or contaminate detection instruments such as gas chromatographs, affecting the performance of the instruments, leading to inaccurate test results or an increase in false positives / false negatives. At the same time, the concentrations of some pesticide residues are low, and existing conventional detection technologies may not be sensitive enough to detect trace amounts of pesticide residues.

[0025] In the above background, the present application takes production raw materials as the starting point for pesticide residue detection. The pesticide residue detection system includes a processing end, which digitally communicates with a first service end, a first detection end, and a second detection end to ensure data interaction. Based on the digital communication between the processing end and the first service end, the first detection end, and the second detection end, the basic data of raw materials used in vinegar production is obtained from the first service end. The basic data of vinegar raw materials includes key information such as the origin and harvest time of the raw materials.

[0026] Through in-depth analysis and processing of the production raw material data set on the processing end, an adaptive similarity clustering algorithm is used to divide raw materials with similar properties into their own similarity cluster sets according to the source location, time and other possible relevant characteristics of the raw materials, and establish similarity cluster sets; according to the preset sampling ratio, an appropriate sample size is determined in each similarity cluster set to ensure that it can represent the characteristics of this type of production raw materials and can achieve effective management under limited resources.

[0027] K2: using the sample size as a collection constraint, controlling the first detection end to perform random sampling of similar cluster sets, and generating a raw material hazard identification according to the random sampling result, and uploading the raw material hazard identification to the first service end, which is then sent to the pesticide residue detection system by the first service end, obtaining monitoring data of raw material synthesis records through the processing end, and predicting vinegar production risks using the monitoring data and the raw material hazard identification to obtain prediction results;

[0028] According to the sample size set in the similar cluster set, the first detection end is controlled to perform random sampling to ensure the representativeness of the sample, and physical sampling and information extraction are performed on the sampled raw materials; the raw materials for vinegar production are marked with raw material hazards according to the random sampling results, and the raw material hazard labels are used to mark the potential pesticide residue risk levels in different vinegar production raw materials, such as safety, low risk, medium risk, and high risk, so as to trigger the corresponding safety mechanism and pay attention to the use of high-risk raw materials;

[0029] The generated raw material hazard identification is uploaded to the first server to facilitate data integration and sharing between systems. After receiving the hazard identification, the first server forwards it to the pesticide residue detection system for subsequent risk analysis and prediction; the processing end obtains the hazard identification of the raw material and its corresponding synthetic record monitoring data (such as raw material batch, production process, etc.) by connecting to the data interface of the pesticide residue detection system; combining the raw material hazard identification with the monitoring data of the production process, the machine learning algorithm is used to predict the potential risks of vinegar production and obtain the prediction results; a series of operations from data collection, similarity clustering, sample selection, pesticide residue detection to final risk prediction are intended to effectively prevent and reduce food safety hazards in the production process through strict monitoring and refined management of raw materials for vinegar production.

[0030] The use of machine learning algorithms to predict potential risks in vinegar production also includes: taking a logistic regression function model that assumes that the dependent variable obeys a logistic distribution, a Cox proportional risk model, a Poisson regression model, or any other risk prediction model as the model basis, using the acquired raw material hazard identification and various monitoring data in the production process to train or apply machine learning models for risk prediction. The risk prediction model will establish association rules by learning historical data, so that it can quantitatively evaluate the potential risks in the vinegar production process and predict possible safety hazards, such as the probability of serious pesticide residue exceeding the standard under a certain combination or condition.

[0031] Furthermore, the raw material hazard identification is generated in K2, and the method of the present application further includes:

[0032] K21: Classify the random sampling results into similar cluster sets as analysis units, and calculate the sampling risk mean of each data classification result to generate an initial risk identification;

[0033] K22: Obtain the maximum difference by calculating the difference between the random sampling result of the corresponding data classification result and the sampling hazard mean calculation result, configure a penalty factor based on the maximum difference, use the penalty factor to penalize and compensate the initial hazard identification, and generate the raw material hazard identification according to the output result.

[0034] Use clustering algorithms (such as K-means, DBSCAN, hierarchical clustering, etc.) to analyze the random sampling results and divide them into multiple categories (cluster sets) based on their intrinsic attribute similarities; for each cluster set, calculate the risk index value of each sample and obtain the average value, which is the sampling risk mean of the cluster set. The sampling risk mean is used as a standard to measure the potential risk level of the corresponding category of raw materials. According to the risk mean corresponding to each cluster set, a risk identification is preliminarily assigned to each group of raw materials;

[0035] Compare the specific hazard index values ​​of each random sampling point within each cluster set with the hazard mean of the cluster set, and calculate the differences respectively. The obtained differences reflect the maximum deviation between the specific hazard index values ​​corresponding to a sample in the current cluster set and the hazard mean; based on the maximum difference found, set a corresponding penalty factor function. The purpose of the penalty factor function is to reflect the difference. The greater the difference, the more significant the impact on the overall hazard assessment; use the penalty factor corresponding to the maximum difference to adjust the relevant initial hazard identification. For example, if there is a maximum difference in a cluster set, its absolute value far exceeds the absolute value corresponding to the maximum difference in other cluster sets, indicating that the risk distribution within the cluster set is uneven, and the overall hazard level of the cluster set should be appropriately adjusted upward, and the overall hazard identification should be appropriately adjusted downward; otherwise, the overall hazard identification should be appropriately adjusted downward; after completing the correction of the hazard identification of all cluster sets, integrate and generate the raw material hazard identification to guide subsequent raw material management.

[0036] Furthermore, if Figure 2 As shown, the K22 generates the raw material hazard identification according to the output result, and the method of the present application further includes:

[0037] The standard deviation of each data classification result is calculated based on the difference calculation result;

[0038] Input the standard deviation into a risk compensation channel and output a discrete compensation result, wherein the risk compensation channel is a compensation channel constructed by big data and is a time series update channel;

[0039] The output result is continuously compensated by the discrete compensation result, and the raw material hazard identification is updated.

[0040] In each cluster set, the risk index values ​​of all random sampling points within it are counted; the standard deviation of each cluster set corresponding to the data classification result is calculated. The standard deviation of each data classification result reflects the degree of dispersion of data points in each category to the risk mean. The larger the standard deviation, the higher the variability between data points, and the potential risk may be more unstable; the calculated standard deviation value of each category is input into the risk compensation channel model constructed in advance through big data training. The risk compensation channel is trained based on empirical data and corresponding time series characteristics, and can dynamically predict and adjust the degree of risk compensation according to the standard deviation of real-time data.

[0041] After receiving the standard deviation input, the risk compensation channel will make predictions based on the time series characteristics and output a discrete compensation result corresponding to the input standard deviation. The risk compensation channel is used to quantify the factors that need additional consideration due to data dispersion. The discrete compensation result output by the risk compensation channel model is used. Further, if the standard deviation is large, it indicates that the risk volatility of the raw materials in this category is high. The output result is further compensated by the discrete compensation result, and then the raw material risk identification is updated. The mean and standard deviation are comprehensively considered, and the raw material risk identification is updated by constructing a risk compensation channel, thereby more accurately reflecting the actual risk status of the raw materials.

[0042] K3: configuring a preset ratio of random samples and weighted samples, and setting a total sampling amount, and configuring sampling parameters through the processing end according to the preset ratio, the total sampling amount and the prediction result;

[0043] Furthermore, the sampling parameters are configured in K3, and the method of the present application also includes:

[0044] K31: Calculate the sampling amount of random samples according to the preset ratio and the total sampling amount, and randomly sample and distribute the vinegar produced in time series to establish random sampling points;

[0045] K32: Calculate the sampling amount of weighted samples according to the preset ratio and the total sampling amount, and calculate the weighted distribution of sampling timing points according to the prediction results and the raw material calling situation under the time-series production to establish weighted sampling points;

[0046] K33: Complete sampling parameter configuration through the random sampling points and the weighted sampling points.

[0047] Confirm the mixed sampling method to be used for pesticide residue detection, that is, the specific proportion of random sampling and weighted sampling (for example, 70% random sampling and 30% weighted sampling), set the sampling quantity for each category or batch, and adjust the weighted sampling part according to the risk factor or other relevant factors of each category or batch; configure the sampling parameters using the preset proportion, the total sampling amount and the predicted result, and write the configured sampling parameters to the processing end to guide the subsequent sampling work.

[0048] Further, the sampling parameters configured in K3 also include: calculating the number of samples required for random samples according to a preset ratio (for example, the percentage in the population) and the total number of samples, and the specific calculation method is: preset ratio * population number = random sample sampling amount; in the entire batch of vinegar produced in the time series, the sampling points are distributed and configured according to the principle of pure randomness to ensure the randomness and representativeness of each sampling point selection; the weight distribution principle of weighted sampling can be determined according to historical data (key nodes in the production process, quality fluctuation nodes), risk level or business needs; the total number of samples collected is determined in comparison with the production scale (for example, a total of 1,000 samples need to be tested); the number of random samples (for example, 700) and the number of weighted samples (for example, 300) are calculated; based on the prediction results and the data on the use of raw materials under the time series production, which time periods or batches are more important are analyzed, and different sampling weights are assigned accordingly; using the prediction results and real-time production data, sampling time series points with different weights are determined, and the sampling time series points are usually sampled before and after important events such as possible quality problems, large fluctuations in output or replacement of raw materials.

[0049] The number and distribution of random sampling points and the number and distribution of weighted sampling points are integrated to form a complete sampling plan parameter. Combined with the specific information of random sampling points and weighted sampling points, all sampling locations are laid out in the entire production sequence, where each sampling location represents a specific node, for example, it can be the node where raw material A enters the factory for vinegar production. Key sampling is carried out at the key stage where pesticide residue problems may occur to form the final sampling parameter configuration. The parameters of random sampling and weighted sampling are formulated separately, and these two types of sampling points are reasonably distributed in the context of sequential production. Finally, the two are combined to ensure comprehensive and targeted coverage of the entire production process, effectively realizing quality monitoring and troubleshooting.

[0050] K4: Controlling the second detection end to perform vinegar production detection through the sampling parameters, and generating a pesticide residue detection result, and sending the pesticide residue detection result to the first service end for storage.

[0051] In the vinegar production process, the second detection end is controlled by the sampling parameters, and the pesticide residue detector is used to obtain the pesticide residue detection result corresponding to each sampling point, and the pesticide residue detection result includes the type of pesticide and the residue amount; all the pesticide residue detection results are summarized and sorted, and the pesticide residue detection results are sent to the first server for storage, which is convenient for later query, analysis and tracing, realizing a closed-loop process from sampling, execution of detection to result reporting, and ensuring the timeliness of pesticide residue monitoring work in the vinegar production process.

[0052] Furthermore, the present application method also includes:

[0053] Conducting attention trigger analysis based on the pesticide residue detection results to establish an attention trigger mechanism for vinegar;

[0054] Controlling the second detection end to perform additional sampling through the attention trigger mechanism;

[0055] The pesticide residue detection results stored in the first server are updated according to the additional sampling results.

[0056] Pesticides widely used in agricultural production may cause a certain amount of pesticide residues in vinegar products if they are not fully degraded or washed away after harvest. Based on this, according to GB 2763-2021 "Standards for Maximum Residue Limits of Pesticides in Foods" and other relevant safety standards, 60% of the allowable limit value of pesticide residues is set as the trigger benchmark, and an attention trigger mechanism for vinegar is established; the pesticide residue test results of vinegar are received and analyzed in real time. When the test results show that the pesticide residue concentration is close to or exceeds the set trigger benchmark, the automatic trigger mechanism is immediately activated. Through the attention trigger mechanism, the second detection end is controlled to perform additional sampling tasks, and additional sampling is performed to further deepen the detection of pesticide residues;

[0057] After receiving the instruction, the second detection terminal prepares additional sample collection containers according to the preset additional sampling plan; conducts more detailed sampling of the same batch or other related batches of vinegar products, obtains additional sampling results, conducts review and testing through the collected additional samples, updates the pesticide residue test results, and obtains more accurate pesticide residue data, providing support for the early degradation or cleaning and removal of pesticides in raw materials with pesticide residues. Based on the attention trigger mechanism of the pesticide residue test results, and through additional sampling and updating of pesticide residue test results, the quality of vinegar products is guaranteed.

[0058] Furthermore, the present application method also includes:

[0059] Determine whether the raw material hazard identification meets a preset hazard threshold;

[0060] When the raw material hazard mark meets the preset hazard threshold, a verification and rejection instruction is triggered;

[0061] The verification and rejection instruction is used to control the first detection end to execute re-inspection, collection and verification of the dangerous set in the similar cluster concentration, and raw material management is performed according to the re-inspection, collection and verification results.

[0062] According to GB 2763-2021 "Standards for Maximum Residue Limits of Pesticides in Food" and other relevant safety standards, in addition to updating the pesticide residue test results, it is also necessary to set the pesticide residue limit value as the preset hazard threshold to determine whether the raw material hazard identification meets the preset hazard threshold;

[0063] When the raw material hazard identification meets the preset hazard threshold, it indicates that there is excessive pesticide residue, triggering a verification and rejection instruction; the verification and rejection instruction is sent to the first detection end, and further re-inspection, collection and verification are carried out on the hazard set in the similar cluster where the raw materials suspected of exceeding the standard are located. The re-inspection may include but is not limited to measures such as re-sampling or increasing the frequency of testing to ensure that the results are accurate; after performing the re-inspection, collection and verification, the first detection end will generate a new test result, which is used to decide whether to confirm that the raw materials do have excessive pesticide residues.

[0064] If the re-inspection results show that the raw materials do contain unacceptable levels of pesticide residues, corresponding management decisions will be made accordingly, which may include but are not limited to stopping the use of the batch of raw materials, isolating and storing them, returning them to the supplier or destroying them, so as to ensure that the safety and quality of the vinegar products meet the specified standards, continuously accumulate re-inspection result data, continuously eliminate mechanisms, and improve the level of raw material management.

[0065] Furthermore, the present application method also includes:

[0066] Based on the raw material hazard identification, cumulative hazard combination evaluation is performed to establish combination constraints;

[0067] The material combination control of the vinegar production process is carried out through the combination constraint.

[0068] Obtain hazard identification data for all raw materials, including but not limited to pesticide residues, heavy metal content, microbial contamination and other hazard indicators; analyze the impact of factors such as raw material source, season, batch, etc. on their hazard identification, and establish a detailed raw material database; determine the weights of different hazard identifications based on the degree of hazard of hazardous substances to food safety; based on the raw material database and the weights of different hazard identifications, superimpose and calculate the hazard identifications of different raw materials according to the weights to obtain a comprehensive hazard score for each raw material combination.

[0069] Set safety thresholds and determine the safety range of material combinations based on comprehensive risk scores; formulate material combination constraint strategies, such as prohibiting the mixed use of raw materials with high risk labels and giving priority to raw materials with low risk labels; formulate corresponding risk reduction measures for material combinations with higher risk scores, such as adding cleaning procedures and extending fermentation time. The cumulative risk combination evaluation method based on raw material risk labels has been implemented in practice, promoting the standardized management of material combinations in the vinegar production process and ensuring product quality.

[0070] In summary, the beneficial effects of the embodiments of the present application are:

[0071] 1. Improving the sensitivity and accuracy of pesticide residue detection will help to promptly detect and eliminate unqualified raw materials and ensure the safety of vinegar products.

[0072] 2. Realize the monitoring of the whole process of vinegar production, and effectively control the risk of pesticide residues from raw material procurement to finished product delivery.

[0073] 3. Through intelligent information systems and advanced data analysis methods, the food traceability capability and risk management level have been improved.

[0074] 4. By using the attention trigger analysis based on the pesticide residue test results, the attention trigger mechanism of vinegar is established; the second detection end is controlled to perform additional sampling through the attention trigger mechanism; the pesticide residue test results stored in the first service end are updated according to the additional sampling results. Based on the attention trigger mechanism of the pesticide residue test results, the quality of the vinegar product is guaranteed through additional sampling and updating of the pesticide residue test results.

[0075] Embodiment 2

[0076] Based on the same inventive concept as the method for detecting pesticide residues in vinegar production in the aforementioned embodiment, Figure 3 As shown, the embodiment of the present application provides a pesticide residue detection system for vinegar production, wherein the pesticide residue detection system includes a processing end, which digitally communicates with a first service end, a first detection end, and a second detection end, and the system includes:

[0077] The data set establishment module M100 is used to establish a production raw material data set of vinegar, wherein the production raw material data set is basic data of vinegar raw materials obtained from the first server end after the pesticide residue detection system establishes digital communication with the first server end, and the production raw material data set includes the source location of the raw materials and the source time of the raw materials. The processing end uses the production raw material data set to perform adaptive similarity clustering of the raw materials, establish a similar cluster set, and set the sample size of the similar cluster set by a preset sampling ratio;

[0078] The random sampling module M200 is used to control the first detection end to perform random sampling of similar cluster sets with the sample size as a collection constraint, and to generate a raw material hazard identification according to the random sampling result, and to upload the raw material hazard identification to the first service end, which is then sent to the pesticide residue detection system by the first service end, and to obtain the monitoring data of the raw material synthesis record through the processing end, and to perform vinegar production risk prediction with the monitoring data and the raw material hazard identification to obtain a prediction result;

[0079] A preset ratio configuration module M300 is used to configure a preset ratio of random samples and weighted samples, and set the total sampling amount, and configure sampling parameters according to the preset ratio, the total sampling amount and the prediction result through the processing end;

[0080] The vinegar production detection module M400 is used to control the second detection end to perform vinegar production detection through the sampling parameters, generate pesticide residue detection results, and send the pesticide residue detection results to the first service end for storage.

[0081] Furthermore, the random sampling module M200 is also used to perform the following steps:

[0082] The random sampling results are classified into data with similar cluster sets as analysis units, and the sampling risk mean calculation of each data classification result is performed to generate an initial risk identification;

[0083] The maximum difference is obtained by calculating the difference between the random sampling result of the corresponding data classification result and the sampling hazard mean calculation result, and a penalty factor is configured based on the maximum difference. The initial hazard identification is penalized and compensated with the penalty factor, and the raw material hazard identification is generated according to the output result.

[0084] Furthermore, the random sampling module M200 is also used to perform the following steps:

[0085] The standard deviation of each data classification result is calculated based on the difference calculation result;

[0086] Input the standard deviation into a risk compensation channel and output a discrete compensation result, wherein the risk compensation channel is a compensation channel constructed by big data and is a time series update channel;

[0087] The output result is continuously compensated by the discrete compensation result, and the raw material hazard identification is updated.

[0088] Furthermore, the preset ratio configuration module M300 is also used to perform the following steps:

[0089] The sampling amount of the random sample is calculated according to the preset ratio and the total sampling amount, and the vinegar produced in time series is randomly sampled and distributed to establish random sampling points;

[0090] The sampling amount of weighted samples is calculated based on the preset ratio and the total sampling amount, and the weighted distribution calculation of sampling timing points is performed based on the prediction results and the raw material calling situation under the time-series production to establish weighted sampling points;

[0091] The sampling parameter configuration is completed through the random sampling points and the weighted sampling points.

[0092] Furthermore, the pesticide residue detection system for vinegar production is also used to perform the following steps:

[0093] Conducting attention trigger analysis based on the pesticide residue detection results to establish an attention trigger mechanism for vinegar;

[0094] Controlling the second detection end to perform additional sampling through the attention trigger mechanism;

[0095] The pesticide residue detection results stored in the first server are updated according to the additional sampling results.

[0096] Furthermore, the pesticide residue detection system for vinegar production is also used to perform the following steps:

[0097] Determine whether the raw material hazard identification meets a preset hazard threshold;

[0098] When the raw material hazard mark meets the preset hazard threshold, a verification and rejection instruction is triggered;

[0099] The verification and rejection instruction is used to control the first detection end to execute re-inspection, collection and verification of the dangerous set in the similar cluster concentration, and raw material management is performed according to the re-inspection, collection and verification results.

[0100] Furthermore, the pesticide residue detection system for vinegar production is also used to perform the following steps:

[0101] Based on the raw material hazard identification, cumulative hazard combination evaluation is performed to establish combination constraints;

[0102] The material combination control of the vinegar production process is carried out through the combination constraint.

[0103] In summary, any step can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor, and no unnecessary restrictions are made here.

[0104] Furthermore, the above-mentioned technical scheme only reflects the preferred technical scheme of the technical scheme of the embodiment of the present application. Some changes that may be made to some parts thereof by technicians in this technical field all reflect the novel principles of the embodiment of the present application. Obviously, technicians in this field can make various changes and modifications to the present application without departing from the scope of the present application.

Claims

1. A method for detecting pesticide residues in vinegar production, characterized in that: The method is applied to a pesticide residue detection system, the pesticide residue detection system includes a processing end, digitally communicating with a first service end, a first detection end, and a second detection end, and the method includes: K1: Establishing a production raw material data set of vinegar, wherein the production raw material data set is basic data of vinegar raw materials obtained from the first server end after the pesticide residue detection system establishes digital communication with the first server end, and the production raw material data set includes the source location of the raw materials and the source time of the raw materials. The processing end uses the production raw material data set to perform adaptive similarity clustering of the raw materials, establish a similar cluster set, and set the sample size of the similar cluster set by a preset sampling ratio; K2: using the sample size as a collection constraint, controlling the first detection end to perform random sampling of similar cluster sets, and generating a raw material hazard identification according to the random sampling result, and uploading the raw material hazard identification to the first service end, which is then sent to the pesticide residue detection system by the first service end, obtaining monitoring data of raw material synthesis records through the processing end, and predicting vinegar production risks using the monitoring data and the raw material hazard identification to obtain prediction results; K3: configuring a preset ratio of random samples and weighted samples, and setting a total sampling amount, and configuring sampling parameters through the processing end according to the preset ratio, the total sampling amount and the prediction result; K4: Controlling the second detection end to perform vinegar production detection through the sampling parameters, and generating a pesticide residue detection result, and sending the pesticide residue detection result to the first service end for storage.

2. The method according to claim 1, characterized in that The generation of raw material hazard identification in K2 also includes: K21: Classify the random sampling results into similar cluster sets as analysis units, and calculate the sampling risk mean of each data classification result to generate an initial risk identification; K22: Obtain the maximum difference by calculating the difference between the random sampling result of the corresponding data classification result and the sampling hazard mean calculation result, configure a penalty factor based on the maximum difference, use the penalty factor to penalize and compensate the initial hazard identification, and generate the raw material hazard identification according to the output result.

3. The method according to claim 2, characterized in that The K22 generates the raw material hazard mark according to the output result, and also includes: The standard deviation of each data classification result is calculated based on the difference calculation result; Input the standard deviation into a risk compensation channel and output a discrete compensation result, wherein the risk compensation channel is a compensation channel constructed by big data and is a time series update channel; The output result is continuously compensated by the discrete compensation result, and the raw material hazard identification is updated.

4. The method according to claim 1, characterized in that The sampling parameters configured in K3 also include: K31: Calculate the sampling amount of random samples according to the preset ratio and the total sampling amount, and randomly sample and distribute the vinegar produced in time series to establish random sampling points; K32: Calculate the sampling amount of weighted samples according to the preset ratio and the total sampling amount, and calculate the weighted distribution of sampling timing points according to the prediction results and the raw material calling situation under the time-series production to establish weighted sampling points; K33: Complete sampling parameter configuration through the random sampling points and the weighted sampling points.

5. The method according to claim 1, characterized in that The method further comprises: Conducting attention trigger analysis based on the pesticide residue detection results to establish an attention trigger mechanism for vinegar; Controlling the second detection end to perform additional sampling through the attention trigger mechanism; The pesticide residue detection results stored in the first server are updated according to the additional sampling results.

6. The method according to claim 1, characterized in that The method further comprises: Determine whether the raw material hazard identification meets a preset hazard threshold; When the raw material hazard mark meets the preset hazard threshold, a verification and rejection instruction is triggered; The verification and rejection instruction is used to control the first detection end to execute re-inspection, collection and verification of the dangerous set in the similar cluster concentration, and raw material management is performed according to the re-inspection, collection and verification results.

7. The method according to claim 1, characterized in that The method further comprises: Based on the raw material hazard identification, cumulative hazard combination evaluation is performed to establish combination constraints; The material combination control of the vinegar production process is carried out through the combination constraint.

8. A pesticide residue detection system for vinegar production, characterized in that: Used to implement the method for detecting pesticide residues in vinegar production according to any one of claims 1 to 7, the pesticide residue detection system comprises a processing end, digitally communicating with a first service end, a first detection end, and a second detection end, the system comprising: A data set establishment module is used to establish a production raw material data set of vinegar, wherein the production raw material data set is basic data of vinegar raw materials obtained from the first server end after the pesticide residue detection system establishes digital communication with the first server end, and the production raw material data set includes the source location of the raw materials and the source time of the raw materials. The processing end uses the production raw material data set to perform adaptive similarity clustering of the raw materials, establish a similar cluster set, and set the sample size of the similar cluster set by a preset sampling ratio; A random sampling module, used to control the first detection end to perform random sampling of similar cluster sets with the sample size as a collection constraint, and to generate a raw material hazard identification according to the random sampling result, and to upload the raw material hazard identification to the first service end, which is then sent to the pesticide residue detection system by the first service end, and to obtain monitoring data of raw material synthesis records through the processing end, and to perform vinegar production risk prediction based on the monitoring data and the raw material hazard identification to obtain a prediction result; A preset ratio configuration module, used to configure a preset ratio of random samples and weighted samples, and set the total sampling amount, and configure sampling parameters according to the preset ratio, the total sampling amount and the prediction result through the processing end; The vinegar production detection module is used to control the second detection end to perform vinegar production detection through the sampling parameters, generate pesticide residue detection results, and send the pesticide residue detection results to the first service end for storage.

Images