Rice noodle production data tracing method and device based on blockchain technology

Abstract

The present invention relates to the field of blockchain storage technology, and in particular to a rice noodle production data traceability method and device based on blockchain technology. The method obtains anomaly metrics through the deviations between rice noodle production data on different production parameters, and obtains production clusters based on the clustering of anomaly metrics between rice noodle production data; in each production cluster, the influence of each type of production parameter on the anomaly metrics is combined with the data aggregation in the production cluster to obtain the data anomaly of each rice noodle production data, and the rice noodle production data is combined for blockchain storage. The present invention combines the high randomness of the abnormal links in the production process, and analyzes the abnormalities of each rice noodle data before blockchain storage, obtains the data anomalies and stores them on the chain together, improves the analysis efficiency of accurate traceability, and makes the management of rice noodle production data safer and more reliable.

Description

Technical Field

[0001] The present invention relates to the field of blockchain storage technology, and in particular to a method and device for tracing rice noodle production data based on blockchain technology. Background Art

[0002] Blockchain technology is a technology that records and transmits information in a decentralized manner. Its core feature is to store data in multiple nodes in a distributed manner and connect these nodes into an unalterable chain, namely blockchain, through encryption technology. Each blockchain node has a copy of the entire chain, so there is no single centralized control point, thereby ensuring the security of the data of the entire chain, and using the characteristics of blockchain to verify and record data in rice noodle production. By analyzing and verifying the rice noodle production data, and combining the unalterability of blockchain to trace the production links of rice noodles, the reliable traceability of rice noodle safety management can be improved.

[0003] Due to the tamper-proof nature of blockchain technology, when tracing the rice noodle production data information chain, it is necessary to traverse and check each data link one by one. This requires reviewing more redundant information during the traceability process, and the screening error rate for abnormal traceability information is high. The efficiency of the traceability process is low, which affects the subsequent analysis and management of rice noodle production data. Summary of the invention

[0004] In order to solve the problem that the efficiency of the traceability process in the prior art is low and affects the subsequent analysis and management of rice noodle production data, the purpose of the present invention is to provide a rice noodle production data traceability method and device based on blockchain technology. The technical solution adopted is as follows:

[0005] The embodiment of the present invention provides a rice noodle production data tracing method based on blockchain technology, the method comprising:

[0006] Obtain each piece of rice noodle production data, each piece of rice noodle production data including different types of production parameters;

[0007] According to the deviation distribution of each two rice noodle production data in different types of production parameters, the abnormality measurement of each two rice noodle production data is obtained; based on the abnormality measurement, the rice noodle production data is clustered to obtain a production cluster;

[0008] According to the abnormal measurement of each rice noodle production data in other rice noodle production data in the production cluster, the influence degree of each type of production parameter on the abnormal measurement, and the aggregation distribution degree of the rice noodle production data in the production cluster, the data abnormality of each rice noodle production data is obtained;

[0009] Each piece of rice noodle production data and the corresponding data anomaly are stored on the blockchain.

[0010] Furthermore, the method for obtaining the abnormality metric includes:

[0011] For any two rice noodle production data, the difference of the production parameters of the two rice noodle production data in each category is used as the parameter difference of the two rice noodle production data in each category;

[0012] Merging all types of parameter differences of the two rice noodle production data to obtain a numerical difference index of the two rice noodle production data;

[0013] According to the confusion degree of all kinds of parameter differences of the two rice noodle production data, the distribution difference index of the two rice noodle production data is obtained;

[0014] The numerical difference index and the distribution difference index of the two rice noodle production data are combined to obtain the abnormal measurement of the two rice noodle production data.

[0015] Furthermore, the method for obtaining the distribution difference index includes:

[0016] The number of the numerical value types of all types of parameter differences of the two rice noodle production data is used as the difference distribution quantity of the two rice noodle production data;

[0017] Calculate the variance of the numerical values ​​of all kinds of parameter differences of the two rice noodle production data to obtain the difference confusion value of the two rice noodle production data;

[0018] The distribution difference index of the two rice noodle production data is obtained by combining the difference distribution quantity and the difference confusion value of the two rice noodle production data.

[0019] Furthermore, the method for obtaining the data anomaly includes:

[0020] According to the anomaly measurement between the rice noodle production data in each production cluster and the distribution quantity of the rice noodle production data, the clustering anomaly possibility of each production cluster is obtained;

[0021] Between each piece of rice noodle production data and other rice noodle production data in the production cluster, according to the degree of change of the abnormal metric and the abnormal metric after the production parameter of each type is removed, the abnormal influence of each piece of rice noodle production data on the production parameter of each type is obtained;

[0022] According to the possibility of clustering anomaly of the production cluster in which each piece of rice noodle production data is located and the degree of abnormal influence on the production parameters of each type, the parameter abnormality of each piece of rice noodle production data in each type of production parameter is obtained;

[0023] The data anomaly of each rice noodle production data is obtained by combining the parameter anomaly of all types of production parameters in each rice noodle production data.

[0024] Furthermore, the possible methods for obtaining the aggregation anomaly include:

[0025] The number of all rice noodle production data in each production cluster is taken as the cluster number of each production cluster; the ratio of the cluster number of each production cluster to the total number of all rice noodle production data is taken as the aggregation index of each production cluster;

[0026] In each production cluster, calculate the mean of the anomaly metrics between the rice noodle production data in pairs to obtain the anomaly distribution index of each production cluster;

[0027] Combine the aggregation index and abnormal distribution index of each production cluster to obtain the abnormal significance index of each production cluster; the aggregation index is negatively correlated with the abnormal significance index, and the abnormal distribution index is positively correlated with the abnormal significance index;

[0028] The ratio of the abnormal significance index of each production cluster to the sum of the abnormal significance indexes of all production cluster clusters is used as the clustering abnormality possibility of each production cluster cluster.

[0029] Furthermore, the method for obtaining the abnormal impact degree includes:

[0030] For any piece of rice noodle production data, calculate the mean of production parameters of each type of all other rice noodle production data except the rice noodle production data in the production cluster where the rice noodle production data is located, and obtain other means of production parameters of each type of the rice noodle production data;

[0031] Based on the abnormal measurement acquisition process, other abnormal measurements between the production parameters of each type of the rice noodle production data and other means are obtained;

[0032] The production parameters of each category are taken as the target category in turn, and the target abnormality measurement of the production parameters of each category except the target category and other means of the rice noodle production data is obtained based on the abnormality measurement acquisition process;

[0033] The difference between other abnormality metrics of the rice noodle production data and the target abnormality metric is used as the abnormal impact of each rice noodle production data in the target category.

[0034] Furthermore, the method for obtaining the parameter abnormality includes:

[0035] The product of the abnormal influence of each type of production parameter in each rice noodle production data and the cluster abnormality possibility of the production cluster cluster in which it is located is calculated and normalized to obtain the parameter abnormality of each type of production parameter in each rice noodle production data.

[0036] Furthermore, the parameter anomaly of all types of production parameters in each rice noodle production data is combined to obtain the data anomaly of each rice noodle production data, including:

[0037] The maximum value of the parameter abnormality of all kinds of production parameters in each rice noodle production data is taken as the abnormal maximum value of each rice noodle production data;

[0038] The mean of the parameter abnormality of all types of production parameters in each rice noodle production data is taken as the abnormal mean of each rice noodle production data;

[0039] The product of the abnormal maximum value and the abnormal mean value of each rice noodle production data is negatively correlated and normalized to obtain the data anomaly of each rice noodle production data.

[0040] Furthermore, storing each piece of rice noodle production data and the corresponding data anomaly on the blockchain includes:

[0041] The rice noodle production data and data anomalies are uploaded to the blockchain in order from largest to smallest data anomalies.

[0042] The present invention also provides a rice noodle production data tracing device based on blockchain technology, comprising:

[0043] A data acquisition module is used to acquire each piece of rice noodle production data, each piece of rice noodle production data contains different types of production parameters;

[0044] A cluster analysis module is used to obtain an abnormality metric of each two rice noodle production data according to the deviation distribution of each two rice noodle production data in different types of production parameters; cluster the rice noodle production data based on the abnormality metric to obtain a production cluster;

[0045] An abnormality analysis module is used to obtain the data abnormality of each rice noodle production data according to the abnormality measurement of other rice noodle production data in the production cluster, the influence of each type of production parameter on the abnormality measurement, and the aggregation distribution degree of the rice noodle production data in the production cluster;

[0046] The data chain module is used to store each rice noodle production data and the corresponding data anomalies on the blockchain.

[0047] The present invention has the following beneficial effects:

[0048] The present invention first clusters based on the abnormal metrics between rice noodle production data, and reflects the approximate situation of production data through the obtained production clustering cluster, so as to evaluate the abnormality of production data through clustering situation, and considers that the single random degree of abnormal data in the rice noodle production link is high, so in each production clustering cluster, the influence of each type of production parameter on the abnormal metric is used to adjust the analysis of abnormal metric, and the data aggregation situation in the clustering cluster is clustered, and the data abnormality of each rice noodle production data is comprehensively obtained from the overall abnormal situation of the clustering cluster and the abnormal influence of the data in the cluster on different production parameters, reflecting the possible degree of abnormality in the production data, and combining the rice noodle production data for blockchain storage. The present invention analyzes the abnormal situation of each rice noodle data before blockchain storage, and combines the high randomness of the abnormal link of the production process to obtain the data abnormal situation and store it together, thereby improving the analysis efficiency of accurate traceability and making the management of rice noodle production data safer and more reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] 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.

[0050] Figure 1 A flowchart of a rice noodle production data tracing method based on blockchain technology provided by one embodiment of the present invention;

[0051] Figure 2 A flow chart of a method for obtaining anomaly metrics provided by an embodiment of the present invention;

[0052] Figure 3 A flow chart of a method for obtaining data anomalies provided by an embodiment of the present invention;

[0053] Figure 4 A structural block diagram of a rice noodle production data tracing device based on blockchain technology provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0054] 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 the rice noodle production data traceability method and device based on blockchain technology proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination 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.

[0055] 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.

[0056] The following is a detailed description of a method and device for tracing rice noodle production data based on blockchain technology provided by the present invention in conjunction with the accompanying drawings.

[0057] See also Figure 1 , which shows a flow chart of a rice noodle production data tracing method based on blockchain technology provided by an embodiment of the present invention, the method comprising the following steps:

[0058] S1: Obtain each piece of rice noodle production data, where each piece of rice noodle production data contains different types of production parameters.

[0059] In the embodiment of the present invention, multiple data related to each rice noodle production process will be recorded. A complete rice noodle production process corresponds to a piece of rice noodle production data, wherein multiple data to be recorded are different production parameters, and the production parameters may include the raw material consumption, production time, inspection time, factory time and multiple quality indicators of the rice noodle production, such as the moisture content, protein content or heavy metal content of the rice noodle. These production data are recorded using blockchain technology, which can facilitate the traceability analysis of rice noodle data.

[0060] S2: Obtain anomaly measurements of each two pieces of rice noodle production data according to the deviation distribution of each two pieces of rice noodle production data in different types of production parameters; cluster the rice noodle production data based on the anomaly measurements to obtain production clusters.

[0061] For similar batches of rice noodles, the closer the multiple production parameters are, the more normal the rice noodle production is, and the less abnormal the situation is. When one or more production parameters in the rice noodle production data are deviated, it means that there may be abnormal situations caused by production, and it can be given priority in the subsequent quality traceability analysis. Therefore, the differences between rice noodle production data are analyzed first, and clustering is performed for abnormal analysis.

[0062] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the abnormality metric is described in Figure 2 , which shows a flow chart of a method for obtaining anomaly metrics provided by an embodiment of the present invention, the method comprising the following steps:

[0063] S201: For any two pieces of rice noodle production data, the difference in production parameters of each type between the two pieces of rice noodle production data is used as the parameter difference in each type between the two pieces of rice noodle production data.

[0064] Each piece of rice noodle production data is composed of multiple production parameters, and the similarity of two production situations is judged by the deviation between multiple types of production parameters. In the embodiment of the present invention, the absolute value of the numerical difference between the two pieces of rice noodle production data in each type of production parameter is taken as the parameter difference, and the deviation of the two production processes in a certain production parameter is reflected through the numerical difference.

[0065] S202: integrating all types of parameter differences of the two pieces of rice noodle production data to obtain numerical difference indicators of the two pieces of rice noodle production data.

[0066] When the deviation of the two rice noodle production data in various production parameters is greater, it means that the two rice noodle production data are higher in the degree of numerical deviation and lower in similarity. Therefore, in the embodiment of the present invention, the cumulative value of all kinds of parameter differences can be used as the numerical difference index to reflect the size of the degree of difference as a whole. In other embodiments of the present invention, the mean of all kinds of parameter anomalies can also be used as the numerical difference index to comprehensively reflect the degree of deviation of various production parameters through the average situation of the difference, which will not be further elaborated herein.

[0067] S203: Obtain distribution difference indicators of the two pieces of rice noodle production data according to the chaotic distribution degrees of all types of parameter differences between the two pieces of rice noodle production data.

[0068] For the rice noodle production process, normal production data are consistent or similar. Therefore, when the parameter difference is low, the numerical change is also simple. On the contrary, when there are more abnormal deviations, the differences between the production parameters will be greater, and the numerical performance is the more chaotic the distribution. Therefore, the differences in production parameters are further analyzed from the chaotic distribution.

[0069] Preferably, in the implementation of the present invention, the number of numerical types of all types of parameter differences of the two rice noodle production data is first taken as the difference distribution quantity of the two rice noodle production data. When the number of numerical types of parameter differences is smaller, the difference situation is simpler. The numerical type is the number of numerical values ​​that appear, and numerical values ​​of the same size are one numerical type.

[0070] Furthermore, the variance of the numerical values ​​of all kinds of parameter differences of the two rice noodle production data is calculated to obtain the difference confusion value of the two rice noodle production data. The deviation and confusion of the parameter difference are reflected by the variance. The larger the variance, that is, the larger the difference confusion value, the more complicated the difference situation is, and the lower the approximation of the two rice noodle production data is.

[0071] Finally, the difference distribution quantity and the difference chaos value of the two rice noodle production data are combined to obtain the distribution difference index of the two rice noodle production data. In an embodiment of the present invention, the product of the difference distribution quantity and the difference chaos value is used as the distribution difference index, and they are combined in the form of product. When the difference distribution quantity and the difference chaos value are larger, the deviation is more complicated, and the difference between the two rice noodle production data is higher. In other embodiments of the present invention, the difference distribution quantity and the difference chaos value can also be combined in the form of accumulation to obtain the distribution difference index, which will not be elaborated here.

[0072] S204: Combining the numerical difference index and the distribution difference index of the two pieces of rice noodle production data, obtaining an abnormality measure of the two pieces of rice noodle production data.

[0073] Through the difference in numerical value and distribution between the two production data, the different abnormalities are comprehensively analyzed. In the embodiment of the present invention, the product of the numerical difference index and the distribution difference index is normalized to obtain the abnormality measurement. The larger the numerical difference index and the distribution difference index, the more significant the deviation is, and the higher the difference between the two production processes. In other embodiments of the present invention, the sum of the numerical difference index and the distribution difference index can also be normalized to obtain the abnormality measurement, and combined in the form of addition, which will not be described in detail here.

[0074] At this point, the deviation analysis between every two rice noodle production data is completed, and the abnormality metric is obtained to judge the approximate situation. Further, the rice noodle production data can be clustered based on the abnormality metric to obtain the production cluster.

[0075] In an embodiment of the present invention, the rice noodle production data is clustered by a clustering algorithm according to the abnormality measurement between the rice noodle production data to obtain production clusters, each production cluster represents a production category, and the rice noodle production data with similar production parameters are divided into one category. The clustering algorithm is a technical means well known to those skilled in the art, such as using an iterative self-organizing clustering algorithm, etc., which is not limited or elaborated here.

[0076] S3: According to the anomaly measurement of each rice noodle production data in the production cluster where the rice noodle production data is located, the degree of influence of each type of production parameter on the anomaly measurement, and the degree of clustered distribution of the rice noodle production data in the production cluster, the data anomaly of each rice noodle production data is obtained.

[0077] When dividing the rice noodle production data, the more abnormal data with large differences will be divided into one category, which will be greatly different from the normal cluster in the cluster. In addition, due to the high randomness of the abnormal data in the production process, that is, the production process only has a very significant abnormal impact on a single production parameter, it may be divided into a relatively normal production cluster in the overall analysis. Therefore, the abnormal situation of rice noodle production data cannot be accurately characterized based only on the production cluster situation.

[0078] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining data abnormality can be found in Figure 3 , which shows a flow chart of a method for obtaining data anomalies provided by an embodiment of the present invention, the method comprising the following steps:

[0079] S301: Obtain the clustering anomaly possibility of each production cluster according to the anomaly measurement between the rice noodle production data in each production cluster and the distribution quantity of the rice noodle production data.

[0080] The more data a cluster contains and the smaller the anomaly measure between the data, the more significant the normality of the rice noodle production data in this cluster is. Therefore, according to the cluster distribution of rice noodle production data, the possibility of anomaly analysis is performed on each production cluster.

[0081] Preferably, in an embodiment of the present invention, the number of all rice noodle production data in each production cluster is first taken as the number of clusters of each production cluster, and the ratio of the number of clusters of each production cluster to the total number of all rice noodle production data is taken as the aggregation index of each production cluster. The proportion of data in the cluster to all data is used to reflect the significant situation of normal cluster characterization. The smaller the aggregation index, the smaller the proportion of data in the production cluster, and the more likely it is that the cluster is composed of relatively abnormal data.

[0082] Furthermore, in each production cluster, the mean of the anomaly measurement between the rice noodle production data is calculated to obtain the anomaly distribution index of each production cluster. The numerical distribution of the anomaly measurement in the production cluster reflects the clustered distribution of the production data. When the anomaly distribution index is smaller, it means that the anomaly measurement of the rice noodle production data in the production cluster is smaller as a whole, the degree of approximation is higher, and the production cluster reflects the normal situation more significantly.

[0083] Furthermore, the aggregation index and abnormal distribution index of each production cluster are combined to obtain the abnormal significance index of each production cluster. The distribution and aggregation of data in the cluster comprehensively reflects the abnormal significance of the production cluster. The aggregation index is negatively correlated with the abnormal significance index, and the abnormal distribution index is positively correlated with the abnormal significance index. When the aggregation index is smaller and the abnormal distribution index is larger, it means that the rice noodle production data in the production cluster are less distributed and the abnormal dissimilarity between them is higher. The greater the abnormal significance, the higher the significance of the abnormality of the cluster.

[0084] It should be noted that positive correlation means that the two variables change in the same direction, such as the dependent variable increases as the independent variable increases, and negative correlation means that the two variables change in opposite directions, such as the dependent variable decreases as the independent variable increases. In the embodiment of the present invention, basic mathematical operations can be used to reflect that the aggregation index is negatively correlated with the abnormal significance index, such as addition or multiplication, and the abnormal distribution index is positively correlated with the abnormal significance index, such as division or negative exponential idempotence, which are not limited or elaborated here.

[0085] Finally, the ratio of the abnormal significance index of each production cluster to the sum of the abnormal significance indexes of all production cluster clusters is used as the aggregation abnormality possibility of each production cluster cluster. The aggregation abnormality possibility is obtained by normalizing the sum value, which reflects the abnormal significance of each production cluster cluster, so as to facilitate the subsequent adjustment through the cluster characterization in the specific abnormality analysis.

[0086] S302: between each piece of rice noodle production data and other rice noodle production data in the production cluster, according to the degree of change of the abnormal metric and the abnormal metric after excluding the production parameters of each type, obtain the abnormal influence of each piece of rice noodle production data on the production parameters of each type.

[0087] Even in relatively normal production clusters, there may be random anomalies in rice noodle production data where there may be certain differences in the production parameters of a single type. During the overall analysis, these data are divided into similar normal clusters due to the low level of anomalies. At this time, the anomalies in the rice noodle production data also need to be analyzed.

[0088] Preferably, in an embodiment of the present invention, for any piece of rice noodle production data, first calculate the mean of production parameters of each type of all other rice noodle production data except the rice noodle production data in the production cluster where the rice noodle production data is located, and obtain other means of production parameters of each type of the rice noodle production data. When analyzing each piece of rice noodle production data, analyze the impact of the differences with other rice noodle production data in the same production cluster, and therefore first analyze the calculation of the means of other rice noodle production data.

[0089] In other embodiments of the present invention, the central rice noodle production data in the production cluster where the rice noodle production data is located may also be obtained, and the difference between the rice noodle production data and the central rice noodle production data may be subsequently analyzed.

[0090] Furthermore, based on the abnormal measurement acquisition process, other abnormal measurements between the production parameters of each type of the rice noodle production data and other means are obtained, and through the abnormal measurement acquisition process in step S2, the deviation distribution of the production parameters of different types of the rice noodle production data and the corresponding other means is obtained.

[0091] In other embodiments of the present invention, when the central rice noodle production data reflects the overall situation of other rice noodle production data, the abnormality measurement of the rice noodle production data and the central rice noodle production data is used as other abnormality measurement. The acquisition method is specifically described in step S2 and will not be repeated here.

[0092] Furthermore, the production parameters of each type are taken as the target type in turn, and the target anomaly measurement of the production parameters of each type of rice noodle production data except the target type and other means is obtained based on the process of obtaining the anomaly measurement. After removing the deviation analysis of the target type, the anomaly measurement is calculated again. The changes before and after can reflect the influence of the target type on the overall anomaly measurement.

[0093] In other embodiments of the present invention, when the overall situation of other rice noodle production data is reflected through the central rice noodle production data, the abnormality measurement calculated by the rice noodle production data and other production parameters of the central rice noodle production data except the target type is used as the target abnormality measurement to reflect the impact changes brought about by the production parameters of the target type, which will not be elaborated here.

[0094] Finally, the difference between other anomaly metrics of the rice noodle production data and the target anomaly metrics is used as the abnormal influence of each rice noodle production data in the target category. The degree of influence is reflected by the change difference. The greater the difference, the higher the influence of the target category on the anomaly metric calculation, and the more significant the anomaly caused by the production parameters of the target category may be.

[0095] S303: Obtain the parameter abnormality of each rice noodle production data in each type of production parameter according to the cluster abnormality possibility of the production cluster in which each rice noodle production data is located and the abnormal influence of each type of production parameter.

[0096] For each type of rice noodle production data, the abnormal impact of each type of production parameter is combined with the abnormal situation of the production cluster, and the abnormal situation of each type of production parameter is comprehensively reflected. Preferably, in an embodiment of the present invention, the product of the abnormal impact of each type of production parameter in each rice noodle production data and the cluster abnormality possibility of the production cluster is calculated and normalized to obtain the parameter abnormality of each type of production parameter of each rice noodle production data. It should be noted that normalization is a technical means well known to those skilled in the art, and the selection of the normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0097] S304: combining the parameter abnormality of all types of production parameters in each piece of rice noodle production data to obtain the data abnormality of each piece of rice noodle production data.

[0098] For each piece of rice noodle production data, the abnormal situation of each piece of rice noodle production data is comprehensively reflected by the abnormal performance of each type of production parameter. Preferably, in an embodiment of the present invention, the maximum value of the parameter abnormality of all types of production parameters in each piece of rice noodle production data is first taken as the maximum abnormality value of each piece of rice noodle production data, reflecting the maximum value reflecting the most significant degree of the abnormal situation.

[0099] Furthermore, the mean of the parameter anomaly of all types of production parameters in each rice noodle production data is taken as the abnormal mean of each rice noodle production data, and the overall abnormal significance of the rice noodle production data is reflected by combining all parameter anomalies.

[0100] Finally, the product of the abnormal maximum value and the abnormal mean of each rice noodle production data is normalized to obtain the data anomaly of each rice noodle production data. The larger the abnormal maximum value and the abnormal mean, the more significant the abnormality of a single type of production parameter or the overall production parameter in the rice noodle production data, so the greater the data anomaly.

[0101] At this point, the analysis of the rice noodle production data that currently needs to be uploaded to the chain has been completed, and the data anomalies of each rice noodle production data have been obtained, reflecting the possible degree of existence of anomalies in the rice noodle production data. The greater the data anomaly, the more significant the anomaly in the rice noodle production data.

[0102] S4: Store each piece of rice noodle production data and the corresponding data anomaly on the blockchain.

[0103] Analyzing abnormal situations before storing on the chain can be understood as a data quality control process before the data is uploaded to the chain, which helps to improve the quality of data stored on the blockchain. In addition, based on data abnormalities, they can be quickly located during the traceability process, thereby improving traceability efficiency.

[0104] In an embodiment of the present invention, the rice noodle production data and data anomalies can be uploaded to the blockchain in the order of data anomalies from large to small, so as to realize data cleaning control before uploading the production data of this batch of rice noodles to the blockchain. When tracing the source, the production data with a higher degree of anomalies can be reviewed first, so as to make the traceability process more centralized and efficient and reduce redundant information analysis.

[0105] It should be noted that, in order to facilitate calculation, all indicator data involved in the calculation in the embodiment of the present invention are preprocessed to eliminate the dimension effect. The specific means of eliminating the dimension effect are technical means well known to those skilled in the art and are not limited here.

[0106] In summary, the present invention first clusters based on the abnormal metrics between rice noodle production data, and reflects the approximate situation of production data through the obtained production clustering cluster, so as to evaluate the abnormality of production data through clustering situation, considering that the single random degree of abnormal data in the rice noodle production link is high, therefore, in each production clustering cluster, the influence of each type of production parameter on the abnormal metric is adjusted, and the data aggregation situation in the clustering cluster is adjusted, and the data abnormality of each rice noodle production data is comprehensively obtained from the overall abnormal situation of the clustering cluster and the abnormal influence of the data in the cluster on different production parameters, reflecting the possible degree of abnormality in the production data, and combining the rice noodle production data for blockchain storage. The present invention analyzes the abnormal situation of each rice noodle data before blockchain storage, combines the high randomness of the abnormal link of the production process, obtains the data abnormal situation and stores it together, improves the analysis efficiency of accurate traceability, and makes the management of rice noodle production data safer and more reliable.

[0107] The embodiment of the present invention also provides a rice noodle production data tracing device based on blockchain technology, please refer to Figure 4 , which shows a structural block diagram of a rice noodle production data tracing device based on blockchain technology provided by an embodiment of the present invention, the device includes:

[0108] A data acquisition module is used to acquire each piece of rice noodle production data, each piece of rice noodle production data contains different types of production parameters;

[0109] A cluster analysis module is used to obtain an abnormality metric of each two rice noodle production data according to the deviation distribution of each two rice noodle production data in different types of production parameters; cluster the rice noodle production data based on the abnormality metric to obtain a production cluster;

[0110] An abnormality analysis module is used to obtain the data abnormality of each rice noodle production data according to the abnormality measurement of other rice noodle production data in the production cluster, the influence of each type of production parameter on the abnormality measurement, and the aggregation distribution degree of the rice noodle production data in the production cluster;

[0111] The data chain module is used to store each rice noodle production data and the corresponding data anomalies on the blockchain.

[0112] It should be noted that the device provided in the above embodiment is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. Since the specific implementation process of a rice noodle production data tracing device based on blockchain technology in this embodiment is the same as the specific implementation process of a rice noodle production data tracing method based on blockchain technology described above, it will not be repeated here.

[0113] 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.

[0114] 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 rice noodle production data tracing method based on blockchain technology, characterized in that: The method comprises: Obtain each piece of rice noodle production data, each piece of rice noodle production data includes different types of production parameters; According to the deviation distribution of each two rice noodle production data in different types of production parameters, the abnormality measurement of each two rice noodle production data is obtained; based on the abnormality measurement, the rice noodle production data is clustered to obtain a production cluster; According to the abnormal measurement of each rice noodle production data and other rice noodle production data in the production cluster, the influence of each type of production parameter on the abnormal measurement, and the aggregation distribution degree of the rice noodle production data in the production cluster, the data abnormality of each rice noodle production data is obtained; Store each piece of rice noodle production data and the corresponding data anomalies on the blockchain; The method of clustering the rice noodle production data based on the abnormality metric to obtain a production clustering cluster comprises: According to the abnormal measurement between rice noodle production data, the rice noodle production data is clustered by clustering algorithm, and the rice noodle production data with similar production parameters are divided into one category to obtain production clusters; The method for obtaining the data abnormality includes: According to the anomaly measurement between the rice noodle production data in each production cluster and the distribution quantity of the rice noodle production data, the clustering anomaly possibility of each production cluster is obtained; Between each piece of rice noodle production data and other rice noodle production data in the production cluster, according to the degree of change of the abnormal metric and the abnormal metric after the production parameter of each type is removed, the abnormal influence of each piece of rice noodle production data on the production parameter of each type is obtained; According to the possibility of clustering anomaly of the production cluster in which each piece of rice noodle production data is located and the degree of abnormal influence on the production parameters of each type, the parameter abnormality of each piece of rice noodle production data in each type of production parameter is obtained; Combining the parameter anomaly of all types of production parameters in each piece of rice noodle production data, the data anomaly of each piece of rice noodle production data is obtained; Possible methods for obtaining the aggregation anomaly include: The number of all rice noodle production data in each production cluster is taken as the cluster number of each production cluster; the ratio of the cluster number of each production cluster to the total number of all rice noodle production data is taken as the aggregation index of each production cluster; In each production cluster, calculate the mean of the anomaly metrics between the rice noodle production data in pairs to obtain the anomaly distribution index of each production cluster; Combine the aggregation index and abnormal distribution index of each production cluster to obtain the abnormal significance index of each production cluster; the aggregation index is negatively correlated with the abnormal significance index, and the abnormal distribution index is positively correlated with the abnormal significance index; The ratio of the abnormal significance index of each production cluster to the sum of the abnormal significance indexes of all production clusters is used as the possibility of cluster abnormality of each production cluster; The method for obtaining the abnormal impact degree includes: For any piece of rice noodle production data, calculate the mean of production parameters of each type of all other rice noodle production data except the rice noodle production data in the production cluster where the rice noodle production data is located, and obtain other means of production parameters of each type of the rice noodle production data; Based on the abnormal measurement acquisition process, other abnormal measurements between the production parameters of each type of the rice noodle production data and other means are obtained; The production parameters of each category are taken as the target category in turn, and the target abnormality measurement of the production parameters of each category except the target category and other means of the rice noodle production data is obtained based on the abnormality measurement acquisition process; The difference between other abnormality metrics of the rice noodle production data and the target abnormality metric is used as the abnormal impact of each rice noodle production data in the target category.

2. According to a rice noodle production data tracing method based on blockchain technology according to claim 1, it is characterized in that: The method for obtaining the abnormality metric includes: For any two rice noodle production data, the difference of the production parameters of the two rice noodle production data in each category is used as the parameter difference of the two rice noodle production data in each category; Merging all types of parameter differences of the two rice noodle production data to obtain a numerical difference index of the two rice noodle production data; According to the confusion degree of all kinds of parameter differences of the two rice noodle production data, the distribution difference index of the two rice noodle production data is obtained; Combining the numerical difference index and the distribution difference index of the two rice noodle production data, obtaining an abnormality measure of the two rice noodle production data; The method for obtaining the distribution difference index includes: The number of the numerical value types of all types of parameter differences of the two rice noodle production data is used as the difference distribution quantity of the two rice noodle production data; Calculate the variance of the numerical values ​​of all kinds of parameter differences of the two rice noodle production data to obtain the difference confusion value of the two rice noodle production data; The distribution difference index of the two rice noodle production data is obtained by combining the difference distribution quantity and the difference confusion value of the two rice noodle production data.

3. According to a rice noodle production data tracing method based on blockchain technology according to claim 1, it is characterized in that: The method for obtaining the parameter abnormality includes: The product of the abnormal influence of each type of production parameter in each rice noodle production data and the cluster abnormality possibility of the production cluster cluster in which it is located is calculated and normalized to obtain the parameter abnormality of each type of production parameter in each rice noodle production data.

4. According to a rice noodle production data tracing method based on blockchain technology according to claim 1, it is characterized in that: The method combines the parameter abnormality of all types of production parameters in each piece of rice noodle production data to obtain the data abnormality of each piece of rice noodle production data, including: The maximum value of the parameter abnormality of all kinds of production parameters in each rice noodle production data is taken as the abnormal maximum value of each rice noodle production data; The mean of the parameter abnormality of all types of production parameters in each rice noodle production data is taken as the abnormal mean of each rice noodle production data; The product of the abnormal maximum value and the abnormal mean value of each rice noodle production data is negatively correlated and normalized to obtain the data anomaly of each rice noodle production data.

5. According to a rice noodle production data tracing method based on blockchain technology according to claim 1, it is characterized in that: The storing of each rice noodle production data and the corresponding data anomaly on the blockchain includes: The rice noodle production data and data anomalies are uploaded to the blockchain in order from largest to smallest data anomalies.

6. A rice noodle production data tracing device based on blockchain technology, characterized in that: include: A data acquisition module is used to acquire each piece of rice noodle production data, each piece of rice noodle production data contains different types of production parameters; A cluster analysis module, for obtaining an abnormality measure of each two rice noodle production data according to the deviation distribution of each two rice noodle production data in different types of production parameters; The rice noodle production data is clustered based on the anomaly metric to obtain production clusters, including: According to the abnormal measurement between rice noodle production data, the rice noodle production data is clustered by clustering algorithm, and the rice noodle production data with similar production parameters are divided into one category to obtain production clusters; An abnormality analysis module is used to obtain the data abnormality of each rice noodle production data according to the abnormality measurement of each rice noodle production data and other rice noodle production data in the production cluster, the influence of each type of production parameter on the abnormality measurement, and the aggregation distribution degree of the rice noodle production data in the production cluster; The method for obtaining the data abnormality includes: According to the anomaly measurement between the rice noodle production data in each production cluster and the distribution quantity of the rice noodle production data, the clustering anomaly possibility of each production cluster is obtained; Between each piece of rice noodle production data and other rice noodle production data in the production cluster, according to the degree of change of the abnormal metric and the abnormal metric after the production parameter of each type is removed, the abnormal influence of each piece of rice noodle production data on the production parameter of each type is obtained; According to the possibility of clustering anomaly of the production cluster in which each piece of rice noodle production data is located and the degree of abnormal influence on the production parameters of each type, the parameter abnormality of each piece of rice noodle production data in each type of production parameter is obtained; Combining the parameter anomaly of all types of production parameters in each piece of rice noodle production data, the data anomaly of each piece of rice noodle production data is obtained; Possible methods for obtaining the aggregation anomaly include: The number of all rice noodle production data in each production cluster is taken as the cluster number of each production cluster; the ratio of the cluster number of each production cluster to the total number of all rice noodle production data is taken as the aggregation index of each production cluster; In each production cluster, calculate the mean of the anomaly metrics between the rice noodle production data in pairs to obtain the anomaly distribution index of each production cluster; Combine the aggregation index and abnormal distribution index of each production cluster to obtain the abnormal significance index of each production cluster; the aggregation index is negatively correlated with the abnormal significance index, and the abnormal distribution index is positively correlated with the abnormal significance index; The ratio of the abnormal significance index of each production cluster to the sum of the abnormal significance indexes of all production clusters is used as the possibility of cluster abnormality of each production cluster; The method for obtaining the abnormal impact degree includes: For any piece of rice noodle production data, calculate the mean of production parameters of each type of all other rice noodle production data except the rice noodle production data in the production cluster where the rice noodle production data is located, and obtain other means of production parameters of each type of the rice noodle production data; Based on the abnormal measurement acquisition process, other abnormal measurements between the production parameters of each type of the rice noodle production data and other means are obtained; The production parameters of each category are taken as the target category in turn, and the target abnormality measurement of the production parameters of each category except the target category and other means of the rice noodle production data is obtained based on the abnormality measurement acquisition process; The difference between other abnormality metrics of the rice noodle production data and the target abnormality metric is used as the abnormal impact of each rice noodle production data in the target category; The data chain module is used to store each rice noodle production data and the corresponding data anomalies on the blockchain.

Images