A data collection method and system in food processing

Abstract

The present invention relates to the technical field of food data collection, and specifically to a data collection method and system in a food processing process. The present invention first obtains the abnormal boiling index of the target product according to the difference in boiling temperature time series data of all monitoring positions of the target product; obtains the abnormal sterilization index of the target product according to the deviation of the target product in the target batch and the abnormal degree of the link corresponding to the target product; and then divides all processed products into normal products and abnormal products, and uses different compression methods to compress and store the corresponding food processing data sets of normal products and abnormal products respectively. The present invention determines abnormal products and normal products by analyzing the abnormal conditions of the production links in the processing process, and adopts targeted compression methods for abnormal products and normal products to ensure the quality of the stored collected data and the efficiency of storing collected data, and improves the accuracy of food abnormality analysis to better ensure food safety.

Description

Technical Field

[0001] The present invention relates to the technical field of food data collection, and in particular to a method and system for collecting data in a food processing process. Background Art

[0002] Food safety is a basic issue concerning life safety. Ensuring food safety is the minimum requirement for food consumption, so it is necessary to monitor the food processing process. In order to strictly control the quality of food production, relevant data of each processing link in the food processing process is usually collected and recorded to detect food abnormalities in a timely manner and take corresponding measures to ensure food safety. Since the amount of data generated during food processing is usually very large, directly storing this data will take up a lot of storage space and increase storage and management costs. Therefore, it is very important to compress and store the data in the food processing process.

[0003] When the prior art compresses food processing process data through a unified compression method, lossy compression can effectively improve compression efficiency but it is difficult to ensure data quality, and lossless compression can ensure data quality but the compression efficiency is low. When compressing food processing process data, the prior art does not fully consider that normal product data has a high repeatability and abnormal product data has a lower repeatability than normal product data. Using the same compression method for abnormal product data and normal product data makes the compression method unreasonable, which makes it difficult to ensure the quality of the stored collected data and the efficiency of storing collected data. Summary of the invention

[0004] In order to solve the problem that the existing technology is difficult to ensure the quality of collected data, resulting in inaccurate analysis of food abnormalities and difficulty in ensuring food safety, the purpose of the present invention is to provide a data collection method and system in a food processing process, and the technical solutions adopted are as follows:

[0005] A method for collecting data in a food processing process, the method comprising:

[0006] Obtaining a food processing data set for each processed product in each production batch; the food processing data set includes cooking temperature time series data of all monitoring locations, as well as temperature values ​​after sterilization and sterilization weight change values;

[0007] Take any production batch as the target batch, and any processed product in the target batch as the target product; obtain the abnormal boiling index of the target product according to the difference of boiling temperature time series data of all monitoring positions of the target product;

[0008] According to the differences in temperature values ​​and sterilization weight change values ​​of all processed products corresponding to the target batch and all production batches, the link abnormality of the target batch is obtained; according to the deviation of the target product in the target batch and the link abnormality corresponding to the target product, the sterilization abnormality index of the target product is obtained;

[0009] Based on the abnormal cooking index and the abnormal sterilization index of the processed products, all processed products are divided into normal products and abnormal products, and different compression methods are used to compress and store the food processing data sets corresponding to the normal products and the abnormal products respectively.

[0010] Furthermore, the method for obtaining the abnormal brewing index includes:

[0011] Among all monitoring positions of the target product, the DTW algorithm is used to calculate the DTW distance of the boiling temperature time series data between every two monitoring positions; the mean of all DTW distances is normalized and used as the boiling abnormality indicator of the target product.

[0012] Furthermore, the method for obtaining the abnormality of a link includes:

[0013] According to the differences between the temperature values ​​and the sterilization weight change values ​​corresponding to any two processed products, clustering all processed products in the target batch to obtain a cluster cluster corresponding to the target batch;

[0014] According to the differences in the clustering characteristics of the target batch and the clusters corresponding to all production batches, the abnormality of the link of the target batch is obtained.

[0015] Furthermore, the method for obtaining the clustering clusters includes:

[0016] A two-dimensional coordinate system is constructed with the sterilization weight change value as the horizontal axis and the temperature value after sterilization as the vertical axis; each processed product in the target batch constitutes each data point in the two-dimensional coordinate system; in the two-dimensional coordinate system, all data points are clustered based on an iterative self-organizing clustering algorithm to obtain a cluster cluster corresponding to the target batch.

[0017] Furthermore, the method for obtaining the abnormality of the link of the target batch according to the difference in the aggregation characteristics of the clusters corresponding to the target batch and all the production batches includes:

[0018] In any production batch, according to the Euclidean distance between the data points in the cluster and the cluster center, the abnormal data points in the data points are screened out, and the abnormal data points in the cluster are removed to obtain the updated cluster corresponding to the cluster;

[0019] The aggregation of all updated clusters in the production batch is integrated to obtain the overall aggregation degree of the production batch;

[0020] According to the difference between the target batch and the overall aggregation performance index corresponding to all production batches, the link abnormality of the target batch is obtained.

[0021] Furthermore, the method for obtaining the overall aggregation degree includes:

[0022] In any updated cluster, the Euclidean distance between each data point and the cluster center is used as the deviation parameter of the data point;

[0023] Calculate the sum of the mean of the deviation parameters of all data points and the preset denominator adjustment value, calculate the ratio of the total number of data points to the sum, and obtain the clustering parameters of the updated cluster clusters; calculate the mean of the clustering parameters of all updated cluster clusters in the production batch to obtain the overall clustering degree of the production batch.

[0024] Furthermore, the method for obtaining the sterilization abnormality index includes:

[0025] Calculate the Euclidean distance between the data point corresponding to the target product and the cluster center of the cluster to which it belongs as the deviation factor of the target product;

[0026] The product of the deviation factor of the target product and the abnormality degree of the link corresponding to the target product is normalized to obtain the sterilization abnormality index of the target product.

[0027] Furthermore, the method for obtaining normal products and abnormal products includes:

[0028] Processed products whose abnormal brewing index is greater than the preset abnormal brewing threshold or whose abnormal sterilization index is greater than the preset abnormal sterilization threshold are regarded as abnormal products; processed products other than abnormal products are regarded as normal products.

[0029] Furthermore, the data compression acquisition method includes:

[0030] The food processing data set corresponding to normal products is subjected to lossy compression; and the food processing data set corresponding to abnormal products is subjected to lossless compression.

[0031] A data collection system for a food processing process comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of a data collection method for a food processing process as described in any one of the above items are implemented.

[0032] The present invention provides a data collection system for a food processing process, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of a data collection method for a food processing process are implemented.

[0033] The present invention has the following beneficial effects:

[0034] First, a production batch is selected as the target batch, and a processed product is selected from the batch as the target product. Considering that the boiling temperature time series data of all monitoring positions reflect the temperature changes of different areas in the boiling equipment as the boiling progresses, in order to ensure the boiling effect, the boiling equipment needs to be evenly heated during the boiling process. According to the difference of the boiling temperature time series data of all monitoring positions of the target product, the boiling abnormality index of the target product is obtained. The boiling abnormality index represents the possibility of abnormality of the target product during the boiling process. The larger the boiling abnormality index, the greater the degree of uneven heating of the target product during the boiling process, indicating that the target product is more likely to have an abnormality during the boiling process. First, by comparing the data difference between the target batch and the processed products corresponding to all production batches, the link abnormality of the target batch is obtained. The link abnormality is used to evaluate the abnormality of the target batch in the sterilization link. Analyze the deviation of the target product in the target batch and the link abnormality corresponding to the target product, and obtain the sterilization abnormality index of the target product. The sterilization abnormality index reflects the possibility of abnormality of the target product during the sterilization process. The larger the sterilization abnormality index, the greater the possibility of abnormality of the target product during the sterilization process. By analyzing abnormal situations in the cooking process and the sterilization process during the processing, abnormal products and normal products are determined, and targeted compression methods are used for abnormal products and normal products to ensure the quality and efficiency of stored collected data, improve the accuracy of food abnormality analysis, and better ensure food safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 A flow chart of a method for collecting data in a food processing process provided by one embodiment of the present invention;

[0037] Figure 2 A flow chart of a method for obtaining a link abnormality degree provided by an embodiment of the present invention;

[0038] Figure 3 A flow chart of a method for obtaining a link abnormality degree provided by an embodiment of the present invention;

[0039] Figure 4 A structural diagram of a data acquisition system in a food processing process provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a data collection method and system in a food processing process proposed by the present invention, its specific implementation, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0041] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0042] The following is a detailed description of a method and system for collecting data in a food processing process provided by the present invention in conjunction with the accompanying drawings.

[0043] The embodiment of the present invention provides a method and system for collecting data in a food processing process. Figure 1 , which shows a flow chart of a method for collecting data in a food processing process provided by an embodiment of the present invention, the method comprising the following steps:

[0044] Step S1: Acquire a food processing data set of each processed product in each production batch; the food processing data set includes the cooking temperature time series data of all monitoring positions, as well as the temperature value after sterilization and the sterilization weight change value.

[0045] The present invention is illustrated by an example of any kind of bechamel food processing process. First, the product container of the processed product is customized according to customer needs, the bechamel raw materials required for each product container are prepared, the raw materials are placed in a boiling device for boiling according to steps, and multiple monitoring positions are set in the boiling device. Each monitoring position is temperature-monitored by an installed temperature sensor. After boiling, the boiled bechamel is dried and poured into the product container of the processed product, the filled product container is placed in a sterilization vehicle and arranged neatly, and the filled product container on the sterilization vehicle is pushed into the sterilization device for sterilization to obtain the sterilized processed product, i.e., the bechamel product. Among them, in the present invention, a bechamel food is a bechamel with the same raw materials and the same flavor; a bechamel food has the same gram weight of the processed product, i.e., a bechamel food has the same corresponding product container.

[0046] A food processing data set of each processed product of a seasoning sauce food in each production batch is obtained from the seasoning sauce monitoring database; the food processing data set includes the cooking temperature time series data of all monitoring positions, as well as the temperature value and sterilization weight change value after sterilization corresponding to the sterilization link. Specifically, during the cooking process of the cooking equipment corresponding to the processed product, the temperature sensor at each monitoring position is used to sample the temperature value according to the preset sampling frequency, each temperature value is used as the cooking temperature value, and the cooking temperature value of each monitoring position is sequentially counted according to the time series to obtain the cooking temperature time series data of each monitoring position of the processed product. The temperature sensor and weight sensor on the sterilization vehicle are used to obtain the temperature value and weight value of the processed product before and after sterilization respectively; wherein, the weight value after sterilization is subtracted from the weight value before sterilization to be used as the sterilization weight change value of the processed product.

[0047] In one embodiment of the present invention, sampling is performed according to a preset sampling frequency, and each sampling is taken as a sampling moment, and the preset sampling frequency is 5 times / second. It should be noted that, in order to facilitate calculations, all index data involved in the calculations in the embodiments of the present invention are subjected to data preprocessing to eliminate the dimension effect. The specific means of removing the dimension effect are technical means well known to those skilled in the art and are not limited here. It should be noted that the present invention is directed to all processed products of the same seasoning sauce in each production batch.

[0048] Step S2: taking any production batch as the target batch, and taking any processed product in the target batch as the target product; obtaining the abnormal boiling index of the target product according to the difference of boiling temperature time series data of all monitoring positions of the target product.

[0049] First, select a production batch as the target batch, and select a processed product from the batch as the target product. Considering that the boiling temperature time series data of all monitoring positions reflect the temperature changes in different areas of the boiling equipment as the boiling progresses, in order to ensure the boiling effect, the boiling equipment needs to be evenly heated during the boiling process. According to the differences in the boiling temperature time series data of all monitoring positions of the target product, the boiling abnormality index of the target product is obtained. The boiling abnormality index represents the possibility of abnormality of the target product during the boiling process. The larger the boiling abnormality index, the greater the degree of uneven heating of the target product during the boiling process, which means that the target product is more likely to have abnormalities during the boiling process.

[0050] Considering that temperature is a key factor affecting the taste of seasoning sauces, it affects the quality of processed products. If the temperature difference at different monitoring positions of the boiling equipment is too large, some sauces may not be fully extracted or converted, thus affecting the quality and effect of the final processed products. Considering that the boiling of seasoning sauces involves the process of heating and stirring, under normal circumstances, the raw materials in the boiling equipment will be evenly stirred so that the temperature values ​​at each monitoring position are close. In abnormal circumstances, insufficient stirring will cause different degrees of temperature change at different monitoring positions. For example, the sauces at certain monitoring positions may heat up very quickly because they are close to the heat source and are not stirred sufficiently, thus deviating far from the overall temperature change. Constructing a boiling abnormality index reflects the possibility of abnormalities occurring in the target product during the boiling process. Since the process of obtaining the boiling abnormality index of all processed products in all production batches is the same, the present invention takes the process of obtaining the boiling abnormality index of the target product as an example. Preferably, in one embodiment of the present invention, the method for obtaining the boiling abnormality index includes:

[0051] In all monitoring positions of the brewing equipment corresponding to the target product, the DTW algorithm is used to calculate the DTW distance of the brewing temperature time series data between every two monitoring positions, and the mean of all DTW distances is normalized as the brewing abnormality indicator of the target product. It should be noted that the use of the DTW (Dynamic Time Warping) algorithm to calculate the DTW distance between sequences is a prior art well known to those skilled in the art and will not be described in detail here. It should be noted that normalization is a technical means well known to those skilled in the art, and normalization can adopt linear normalization, etc., which is not limited here.

[0052] Regarding the above steps, considering that the DTW algorithm is a technology used to measure the similarity between two time series, it can handle the expansion and contraction of time series on the time axis, so it is suitable for analyzing the differences in temperature time series data during the brewing process. The DTW distance of the brewing temperature time series data at the two monitoring locations reflects the difference in brewing temperature between local areas in the brewing equipment. The larger the DTW distance, the more likely it is that the abnormal situation of uneven heating will occur during the brewing process. The mean of all DTW distances is calculated to obtain the brewing anomaly index of the target product. The brewing anomaly index reflects the uneven degree of temperature distribution during the brewing process from an overall perspective. The larger the brewing anomaly index, the greater the degree of uneven heating of the target product during the brewing process, which means that the target product is more likely to have an abnormality during the brewing process.

[0053] At this point, the abnormal brewing index of the target product is obtained.

[0054] Step S3: According to the differences in temperature values ​​and sterilization weight change values ​​of all processed products corresponding to the target batch and all production batches, the link abnormality of the target batch is obtained; according to the deviation of the target product in the target batch and the link abnormality corresponding to the target product, the sterilization abnormality index of the target product is obtained.

[0055] Considering that all processed products need to be placed in the sterilization equipment, undergo high-temperature sterilization and cooling before leaving the warehouse. Since the sterilization conditions of the same production batch are the same, ideally the temperature value and sterilization weight change value after sterilization in the same production batch should be relatively close. However, in actual operation, there will be abnormal phenomena of water entering the product container of the processed product. Abnormal phenomena can be divided into two categories: occasional abnormalities and overall abnormalities. Occasional abnormalities: Due to the sealing defects of a single product container, water enters the product container during the sterilization process, thereby changing the temperature value and weight of the processed product, causing a small part of the processed products in the production batch to deviate from the overall distribution of the production batch. Overall abnormality: Due to the aging of the sterilization equipment on the production line or improper parameter settings, the products in the entire production batch have incomplete cooling or different degrees of water entering the bottle, resulting in a large difference in the distribution of all processed products in the production batch. First, by comparing the data differences between the target batch and the corresponding processed products of all production batches, the link abnormality of the target batch is obtained. The link abnormality is used to evaluate the abnormality of the target batch in the sterilization link. Analyze the deviation of the target product in the target batch and the abnormality of the corresponding link of the target product to obtain the sterilization abnormality index of the target product. The sterilization abnormality index reflects the possibility of abnormality of the target product during the sterilization process. The larger the sterilization abnormality index, the greater the possibility of abnormality of the target product during the sterilization process.

[0056] Since the process of obtaining the abnormal sterilization index of all processed products in all production batches is the same, the present invention takes the process of obtaining the abnormal sterilization index of the target product as an example for illustration.

[0057] To assess the degree of sterilization anomalies in the target batch, refer to Figure 2 , which shows a flow chart of a method for obtaining a link abnormality degree in one embodiment of the present invention. Preferably, in one embodiment of the present invention, the method for obtaining a link abnormality degree includes:

[0058] Step S301: clustering all processed products in a target batch according to the differences in the post-sterilization temperature values ​​and sterilization weight change values ​​corresponding to any two processed products, and obtaining clusters corresponding to the target batch.

[0059] According to the difference in temperature values ​​and sterilization weight change values ​​corresponding to any two processed products, all processed products in the target batch are clustered. The purpose of clustering is to classify processed products with similar sterilization temperature values ​​and similar sterilization weight change values ​​into one category to form clusters corresponding to similar sterilization treatments. Preferably, in one embodiment of the present invention, the method for obtaining clusters includes:

[0060] A two-dimensional coordinate system is constructed with the sterilization weight change value as the horizontal axis and the temperature value after sterilization as the vertical axis; each processed product in the target batch constitutes each data point in the two-dimensional coordinate system; in the two-dimensional coordinate system, cluster analysis is performed on the corresponding data points of all processed products based on the iterative self-organizing clustering algorithm to obtain the cluster clusters corresponding to the target batch. It should be noted that the specific process of the iterative self-organizing clustering algorithm is a technical means well known to those skilled in the art, and the various parameters required in the clustering process can be adjusted according to the implementation scenario, which is not limited or elaborated here. In other embodiments of the present invention, the K-means clustering algorithm is also used to perform cluster analysis on the corresponding data points of all processed products in the two-dimensional coordinate system, wherein the K-means clustering algorithm is a technical means well known to those skilled in the art, which is not elaborated here.

[0061] Step S302: Obtain the link abnormality degree of the target batch according to the difference in aggregation characteristics between the target batch and the clusters corresponding to all production batches.

[0062] Considering that the overall abnormality in the production batch is due to the aging of the sterilization equipment on the production line or improper parameter settings, the entire production batch is abnormal, that is, the products in the production batch are not completely cooled or water enters the bottle to varying degrees, which will make the data points in the clusters corresponding to the production batch more scattered; sporadic abnormalities are due to the sealing defects of a single product container, which causes water to enter the product container during the sterilization process, thereby changing the temperature value and weight of the processed product, causing a small number of processed products to deviate from the overall distribution of the production batch. Since the production process needs to ensure the quality of sterilization, the clustering characteristics of the clusters corresponding to all production batches reflect the data aggregation characteristics when the sterilization quality is good. According to the difference in the aggregation characteristics of the clusters corresponding to the target batch and all production batches, the link abnormality of the target batch is obtained. The greater the link abnormality, the worse the aggregation degree of the target batch compared with the clusters corresponding to all production batches, indicating that the sterilization link of the target batch is more likely to be an abnormal link.

[0063] See also Figure 3 , which shows a flow chart of a method for obtaining link abnormality in one embodiment of the present invention. Preferably, in one embodiment of the present invention, according to the difference in clustering characteristics between the target batch and the clusters corresponding to all production batches, the method for obtaining the link abnormality of the target batch includes:

[0064] Step S311: In any production batch, according to the Euclidean distance between the data points in the cluster and the cluster center, the abnormal data points in the data points are screened out, and the abnormal data points in the cluster are removed to obtain the updated cluster corresponding to the cluster.

[0065] In order to analyze the clustering characteristics of production batches, considering the interference of occasional anomalies on the overall anomaly analysis, the abnormal data points reflect the processed products corresponding to the sealing defects. The abnormal data points in the clusters are removed to obtain the updated clusters corresponding to the clusters.

[0066] Specifically, the Euclidean distance between each data point and the cluster center is normalized to obtain the deviation value of the data point; the data point with a deviation value greater than the preset deviation threshold is regarded as an abnormal data point, and the abnormal data point in the cluster is removed to obtain the updated cluster corresponding to the cluster. It should be noted that the sterilization abnormality index of the processed product corresponding to the abnormal data point is set to 1, indicating that there is an abnormality in the sterilization process of the processed product. In one embodiment of the present invention, the preset deviation threshold is 0.47, and the implementer can set it according to the implementation scenario.

[0067] Step S312: The aggregation conditions of all updated clusters in the production batch are integrated to obtain the overall aggregation degree of the production batch.

[0068] Considering that sporadic anomalies are caused by sealing defects in a single product container, water enters the product container during the sterilization process, thereby changing the temperature and weight of the processed product, causing a small number of processed products to deviate from the overall distribution of the production batch, that is, the clusters corresponding to the production batch show obvious clustering characteristics after excluding some abnormal data points. The overall aggregation degree reflects the overall aggregation degree of the production batch after excluding sporadic anomalies. The larger the overall aggregation degree value, the greater the overall aggregation degree after excluding sporadic anomalies.

[0069] Preferably, in one embodiment of the present invention, the method for obtaining the overall aggregation degree includes:

[0070] In any updated cluster, the Euclidean distance between each data point and the cluster center is used as the deviation parameter of the data point;

[0071] Calculate the sum of the mean of the deviation parameters of all data points and the preset denominator adjustment value, calculate the ratio of the total number of data points to the sum, and obtain the clustering parameters of the updated cluster clusters; calculate the mean of the clustering parameters of all updated cluster clusters in the production batch, and obtain the overall clustering degree of the production batch. In one embodiment of the present invention, the preset denominator adjustment value is used to ensure that the denominator is not 0, and the preset denominator adjustment value is a number greater than 0. This embodiment sets it to 0.001, and the implementer can set it according to the implementation scenario.

[0072] For the above steps, when the data points in the updated cluster are closer to the cluster center and the number is larger, it means that the degree of aggregation of the updated cluster is higher, and the aggregation parameter is used to reflect the degree of aggregation of the updated cluster. Calculate the mean of the aggregation parameters of all updated clusters in the production batch to obtain the overall degree of aggregation of the production batch. The larger the value of the overall degree of aggregation, the more aggregated the products in the production batch are after excluding occasional anomalies, and the less likely the sterilization link is an abnormal link.

[0073] Step S313: Obtain the link abnormality of the target batch according to the difference between the target batch and the overall aggregation performance index corresponding to all production batches.

[0074] The greater the abnormality of the link, the worse the clustering degree of the target batch compared with the clustering clusters corresponding to all production batches, which means that the sterilization link of the target batch is likely to be an abnormal link.

[0075] In one embodiment of the present invention, the mean of the overall aggregation performance index corresponding to all production batches is calculated as the reference aggregation index, and the negative correlation mapping result of the difference between the overall aggregation performance index of the target batch and the reference aggregation index is calculated to obtain the link abnormality of the target batch. It should be noted that negative correlation mapping is a technical means well known to those skilled in the art, and negative correlation mapping can be in the form of inverse proportion or negative exponential power without limitation.

[0076] In order to analyze the possibility of abnormality of the target product during the sterilization process, in one embodiment of the present invention, a method for obtaining a sterilization abnormality index includes:

[0077] The Euclidean distance between the data point corresponding to the target product and the cluster center of the cluster to which it belongs is calculated as the deviation factor of the target product; the product of the deviation factor of the target product and the abnormal degree of the link corresponding to the target product is normalized to obtain the sterilization abnormality index of the target product. It should be noted that normalization is a technical means well known to those skilled in the art, and normalization can be performed by linear normalization, etc., which is not limited here.

[0078] For the above steps, the larger the deviation factor, the greater the degree of deviation of the target product from the overall production batch, the more likely the target product is to have sterilization abnormalities, the greater the abnormality of the link corresponding to the target product, the greater the possibility that the sterilization link of the target product is an abnormal link, and the target product's deviation factor and the link abnormality corresponding to the target product are combined to obtain the target product's sterilization abnormality index. The sterilization abnormality index more comprehensively reflects the possibility of abnormalities in the sterilization process of the target product. The larger the sterilization abnormality index, the greater the possibility of abnormalities in the sterilization process of the target product.

[0079] At this point, the sterilization abnormality index of the target product is obtained.

[0080] Step S4: Based on the abnormal cooking index and the abnormal sterilization index of the processed products, all processed products are divided into normal products and abnormal products, and the corresponding food processing data sets of normal products and abnormal products are compressed and stored respectively using different compression methods.

[0081] In order to ensure the data quality of the food processing process, the present invention determines abnormal products and normal products by analyzing the abnormal conditions of the cooking process and the sterilization process during the processing, and adopts targeted compression methods for the abnormal products and normal products to ensure the quality of the stored collected data and the efficiency of storing collected data, improve the accuracy of food abnormality analysis to better ensure food safety.

[0082] Considering that the processed products are mainly affected by the defects in the brewing process and the sterilization process, resulting in abnormal products, and considering that the brewing abnormality index represents the possibility that the target product will have an abnormality in the brewing process, the larger the brewing abnormality index, the greater the possibility that the target product will have an abnormality in the brewing process, and the more likely the processed product is an abnormal product. The sterilization abnormality index reflects the possibility that the target product will have an abnormality in the sterilization process. The larger the sterilization abnormality index, the greater the possibility that the target product will have an abnormality in the sterilization process, and the more likely the processed product is an abnormal product. Preferably, in one embodiment of the present invention, the method for obtaining normal products and abnormal products includes:

[0083] Processed products whose abnormal brewing index is greater than the preset abnormal brewing threshold or whose abnormal sterilization index is greater than the preset abnormal sterilization threshold are regarded as abnormal products; processed products other than abnormal products are regarded as normal products. In one embodiment of the present invention, the preset abnormal brewing threshold is 0.37, and the preset abnormal sterilization threshold is 0.42, which can be set by the implementer according to the implementation scenario.

[0084] Preferably, in one embodiment of the present invention, the method for obtaining data compression includes:

[0085] The food processing data set corresponding to normal products is subjected to lossy compression; the food processing data set corresponding to abnormal products is subjected to lossless compression, and the food processing data set of each processed product in each production batch is subjected to data compression. It should be noted that lossy compression and lossless compression are prior arts well known to those skilled in the art and will not be described in detail here.

[0086] Specifically, the data is stored after compression, and subsequent staff decompresses the data as needed, so that the staff can timely screen abnormal products and maintain defective processing links.

[0087] The present invention also proposes a data acquisition system in the food processing process, please refer to Figure 4 , which shows a structural diagram of a data acquisition system in a food processing process provided by an embodiment of the present invention. The system includes: a data acquisition module 101, a cooking abnormality analysis module 102, a sterilization abnormality analysis module 103 and a compression storage module 104.

[0088] The data acquisition module 101 is used to acquire a food processing data set of each processed product in each production batch; the food processing data set includes the cooking temperature time series data of all monitoring positions, as well as the temperature value after sterilization and the sterilization weight change value.

[0089] The brewing abnormality analysis module 102 is used to take any production batch as the target batch and any processed product in the target batch as the target product; and obtain the brewing abnormality index of the target product according to the difference in the brewing temperature time series data of all monitoring positions of the target product.

[0090] The sterilization abnormality analysis module 103 is used to obtain the link abnormality of the target batch based on the differences in temperature values ​​and sterilization weight change values ​​of all processed products corresponding to the target batch and all production batches; and to obtain the sterilization abnormality index of the target product based on the deviation of the target product in the target batch and the link abnormality corresponding to the target product.

[0091] The compression storage module 104 is used to divide all processed products into normal products and abnormal products based on the abnormal cooking index and sterilization index of the processed products, and use different compression methods to compress and store the corresponding food processing data sets of normal products and abnormal products respectively.

[0092] It should be noted that: the system provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the data collection system in the food processing process provided in the above embodiment and the data collection method embodiment in the food processing process belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0093] In summary, the embodiments of the present invention provide a data collection method and system in the food processing process. First, according to the difference of the cooking temperature time series data of all monitoring positions of the target product, the cooking abnormality index of the target product is obtained; according to the deviation of the target product in the target batch and the abnormality of the link corresponding to the target product, the sterilization abnormality index of the target product is obtained; based on the cooking abnormality index and the sterilization abnormality index of the processed product, all processed products are divided into normal products and abnormal products, and different compression methods are used to compress and store the corresponding food processing data sets of normal products and abnormal products. The present invention determines abnormal products and normal products by analyzing the abnormal conditions of the production links in the processing process, and adopts targeted compression methods for abnormal products and normal products to ensure the quality of the stored collected data and the efficiency of storing collected data, improve the accuracy of food abnormality analysis to better ensure food safety.

[0094] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0095] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A data collection method in a food processing process, characterized in that: The method comprises: Obtaining a food processing data set for each processed product in each production batch; the food processing data set includes cooking temperature time series data of all monitoring locations, as well as temperature values ​​after sterilization and sterilization weight change values; Take any production batch as the target batch, and any processed product in the target batch as the target product; obtain the abnormal boiling index of the target product according to the difference of boiling temperature time series data of all monitoring positions of the target product; According to the differences in temperature values ​​and sterilization weight change values ​​of all processed products corresponding to the target batch and all production batches, the link abnormality of the target batch is obtained; according to the deviation of the target product in the target batch and the link abnormality corresponding to the target product, the sterilization abnormality index of the target product is obtained; Based on the abnormal cooking index and the abnormal sterilization index of the processed products, all processed products are divided into normal products and abnormal products, and the normal products and abnormal products are compressed and stored respectively using different compression methods corresponding to the food processing data sets; The method for obtaining the abnormality of the link includes: Clustering all processed products in the target batch according to the differences in the post-sterilization temperature values ​​and sterilization weight change values ​​corresponding to any two processed products to obtain cluster clusters corresponding to the target batch; According to the difference in the aggregation characteristics of the clusters corresponding to the target batch and all production batches, the abnormality of the link of the target batch is obtained; The method for obtaining the abnormality of the link of the target batch according to the difference in the aggregation characteristics of the clusters corresponding to the target batch and all the production batches includes: In any production batch, according to the Euclidean distance between the data points in the cluster and the cluster center, the abnormal data points in the data points are screened out, and the abnormal data points in the cluster are removed to obtain the updated cluster corresponding to the cluster; The aggregation of all updated clusters in the production batch is integrated to obtain the overall aggregation degree of the production batch; According to the difference between the target batch and the overall aggregation performance index corresponding to all production batches, the link abnormality of the target batch is obtained.

2. The data collection method in a food processing process according to claim 1, characterized in that: The method for obtaining the abnormal brewing index includes: Among all monitoring positions of the target product, the DTW algorithm is used to calculate the DTW distance of the boiling temperature time series data between every two monitoring positions; the mean of all DTW distances is normalized and used as the boiling abnormality indicator of the target product.

3. The data collection method in a food processing process according to claim 1, characterized in that: The method for obtaining the clusters includes: A two-dimensional coordinate system is constructed with the sterilization weight change value as the horizontal axis and the temperature value after sterilization as the vertical axis; each processed product in the target batch constitutes each data point in the two-dimensional coordinate system; in the two-dimensional coordinate system, all data points are clustered based on an iterative self-organizing clustering algorithm to obtain a cluster cluster corresponding to the target batch.

4. The data collection method in a food processing process according to claim 1, characterized in that: The method for obtaining the overall aggregation degree includes: In any updated cluster, the Euclidean distance between each data point and the cluster center is used as the deviation parameter of the data point; Calculate the sum of the mean of the deviation parameters of all data points and the preset denominator adjustment value, calculate the ratio of the total number of data points to the sum, and obtain the clustering parameters of the updated cluster clusters; calculate the mean of the clustering parameters of all updated cluster clusters in the production batch to obtain the overall clustering degree of the production batch.

5. The data collection method in a food processing process according to claim 1, characterized in that: The method for obtaining the sterilization abnormality index includes: Calculate the Euclidean distance between the data point corresponding to the target product and the cluster center of the cluster to which it belongs as the deviation factor of the target product; The product of the deviation factor of the target product and the abnormality degree of the link corresponding to the target product is normalized to obtain the sterilization abnormality index of the target product.

6. The method for data collection in a food processing process according to claim 1, characterized in that: The method for obtaining the normal product and the abnormal product comprises: Processed products whose abnormal brewing index is greater than the preset abnormal brewing threshold or whose abnormal sterilization index is greater than the preset abnormal sterilization threshold are regarded as abnormal products; processed products other than abnormal products are regarded as normal products.

7. The method for data collection in a food processing process according to claim 1, characterized in that: The data compression acquisition method includes: The food processing data set corresponding to normal products is subjected to lossy compression; and the food processing data set corresponding to abnormal products is subjected to lossless compression.

8. A data acquisition system in a food processing process, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the data collection method in a food processing process as described in any one of claims 1 to 7 are implemented.

Images