A two-way traceability method and system for the steel industry chain

Abstract

The present invention discloses a bidirectional traceability method and system for a steel industry chain, comprising the following steps: step S1, sequentially acquiring behavioral feature data representing enterprise behavioral features at multiple enterprise nodes in the steel industry chain, marking the enterprise nodes with traceability authority levels, and mapping the behavioral feature data and traceability authority levels of the enterprise nodes into a traceability model sample sequence; step S2, using a BP neural network to perform model training on the traceability model sample sequence to obtain a traceability model; step S3, identifying the traceability authority levels of enterprise nodes to be identified in the steel industry chain based on the traceability model, and controlling the enterprise nodes to perform bidirectional traceability in the steel industry chain based on the traceability authority levels. The present invention controls the enterprise nodes to perform bidirectional traceability in the steel industry chain based on the traceability authority levels, so as to reduce the amount of traceability data retained in the traceability process based on the centralization of traceability behavior.

Description

Technical Field

[0001] The present invention relates to the field of industrial traceability technology, and in particular to a two-way traceability method and system for the steel industry chain. Background Art

[0002] In daily life, consumers are often victimized by counterfeit and inferior products because it is difficult for them to distinguish genuine products from inferior ones. Counterfeiting is a frequent occurrence, and it is difficult for ordinary consumers to distinguish the difference based on appearance alone. Currently, the market is in disarray, with the amount of "branded agricultural products" available to consumers far exceeding the actual production volume, and product pricing is confusing, making it difficult for consumers to distinguish between genuine and counterfeit products, seriously affecting the growth of brand value and reputation.

[0003] CN202110431930.2 A blockchain-based safe and trusted traceability system for agricultural products, comprising three parts: an Internet of Things system, a video surveillance system, and a decentralized traceability system; the video surveillance system and the Internet of Things system are used to monitor information on key links of agricultural products; the decentralized traceability system is used to verify the consensus of alliance participants on the Internet of Things perception information and video surveillance data information obtained from key links of agricultural products and upload them to the chain. The present invention proposes a strong trust traceability application system that integrates agricultural Internet of Things, intelligent panoramic monitoring, intelligent anti-counterfeiting, and blockchain technology, uploads key life cycle information of branded agricultural products to the chain, and builds a strong trust endorsement tracking and tracing capability for key links of the entire industrial chain of agricultural products that cannot be tampered with or forged. The solution of the present invention utilizes the strong security encryption mechanism of blockchain and the characteristics of open, transparent, and tamper-proof records to overcome many shortcomings of centralized traceability systems.

[0004] Although existing technologies can achieve decentralized traceability, ensuring that all companies have equal traceability rights will cause a large degree of concurrency and retention of data in the traceability system, resulting in a heavy burden on the system and low traceability efficiency. Summary of the Invention

[0005] The purpose of the present invention is to provide a two-way traceability method and system for the steel industry chain to solve the technical problems in the prior art that cause large-scale concurrency and retention of data in the traceability system, resulting in heavy system burden and low traceability efficiency.

[0006] In order to solve the above technical problems, the present invention specifically provides the following technical solutions:

[0007] A two-way traceability method for the steel industry chain includes the following steps:

[0008] Step S1: sequentially obtain behavioral feature data representing enterprise behavioral features at multiple enterprise nodes in the steel industry chain, mark the enterprise nodes with traceability authority levels based on the behavioral feature data, and map the behavioral feature data and traceability authority levels of the enterprise nodes into a traceability model sample sequence;

[0009] Step S2: Using a BP neural network to train the traceability model sample sequence to obtain a traceability model that identifies the traceability authority level of enterprise nodes, so as to quantitatively map the traceability authority level to the enterprise behavior characteristics and realize the centralization of traceability behavior based on the enterprise behavior characteristics;

[0010] Step S3: Based on the traceability model, the traceability authority level of the enterprise nodes to be identified in the steel industry chain is identified, and the enterprise nodes are controlled to perform two-way traceability in the steel industry chain according to the traceability authority level, so as to achieve centralization of traceability behavior to reduce the retention of traceability data in the traceability behavior process.

[0011] As a preferred solution of the present invention, the enterprise behavior characteristics include historical production behavior characteristics and historical traceability behavior characteristics, and the behavior characteristic data include historical production behavior characteristic data and historical traceability behavior characteristic data.

[0012] As a preferred solution of the present invention, the tracing authority level marking of the enterprise node based on the behavioral feature data includes:

[0013] The traceability weights of the historical production behavior feature data and the historical traceability behavior feature data are set, and the historical production behavior feature data and the historical traceability behavior feature data are weighted and summed based on the traceability weights to obtain the traceability authority score of the enterprise node. The calculation formula of the traceability authority score is:

[0014] score i =α i X i +β i Y i ;

[0015] In the formula, score i Represented as the traceability authority score of the i-th enterprise node, X i 、Y i Represented as the historical production behavior feature data and historical traceability behavior feature data of the i-th enterprise node, α i , β i They are respectively represented as the traceability weights of the historical production behavior feature data and the historical traceability behavior feature data of the i-th enterprise node;

[0016] Perform inclusion matching on the traceability authority score and the score confidence interval of the traceability authority level, and mark the traceability authority level corresponding to the score confidence interval containing the traceability authority score as the traceability authority level of the enterprise node;

[0017] Preferably, the setting of the traceability weight includes:

[0018] The total data volume and total duration of the production behavior characteristic data and the historical tracing behavior characteristic data are counted respectively, and the average real-time data flow of the production behavior characteristic data and the historical tracing behavior characteristic data is calculated based on the total data volume and the total duration;

[0019] Normalizing the average real-time data stream of the production behavior characteristic data and the historical traceability behavior characteristic data to obtain the traceability weights of the production behavior characteristic data and the historical traceability behavior characteristic data;

[0020] The calculation formula of the traceability weight is:

[0021]

[0022] Where n i,X 、n i,Y The total amount of production behavior feature data and historical traceability behavior feature data of the i-th enterprise node, t i,X , t i,Y They are respectively represented as the total duration of the production behavior feature data and historical traceability behavior feature data of the i-th enterprise node, It is represented as the average real-time data stream of the production behavior feature data and historical traceability behavior feature data of the i-th enterprise node.

[0023] As a preferred solution of the present invention, the mapping of the behavioral feature data and traceability authority level of the enterprise node into a traceability model sample sequence includes:

[0024] Map the behavior feature data and traceability authority level of the enterprise node into a sequence {X i ,Y i ,level i}, the sequence {X i ,Y i ,level i} as the sample sequence of the traceability model;

[0025] Among them, level i It represents the traceability authority level of the i-th enterprise node, i∈[1,m], m represents the total number of enterprise nodes, and i is a measurement constant.

[0026] As a preferred solution of the present invention, a BP neural network is used to train the traceability model sample sequence to obtain a traceability model for identifying the traceability authority level of enterprise nodes, including:

[0027] The traceability model sample sequence {X i ,Y i ,level i {X in} i ,Y i} as the input of BP neural network, and level i As the output item of the BP neural network, the BP neural network is trained based on the input item and the output item to obtain a traceability model representing the mapping relationship between the behavior feature data and the traceability authority level;

[0028] The expression of the traceability model is:

[0029] level = BP(X,Y);

[0030] Among them, level represents the traceability authority level, {X, Y} represents the behavioral feature data, and BP represents the BP neural network.

[0031] As a preferred solution of the present invention, the method of identifying the traceability authority level of enterprise nodes to be identified in the steel industry chain based on the traceability model includes:

[0032] Extract the behavioral feature data of the enterprise node whose traceability authority level is to be identified in sequence {X j ,Y j}, and the behavioral characteristic data {X j ,Y j Input into the traceability model to obtain the traceability authority level of the enterprise node to be identified j , where X j 、Y j They are respectively represented as the production behavior feature data and historical traceability behavior feature data of the enterprise node of the jth traceability authority level to be identified, level j It is represented as the traceability authority level of the enterprise node with the jth traceability authority level to be identified.

[0033] As a preferred solution of the present invention, the control of enterprise nodes to perform bidirectional traceability in the steel industry chain based on traceability authority levels includes:

[0034] Calculate the bidirectional traceability authority of the enterprise node based on the traceability authority level and traceability authority score of the enterprise node, and perform bidirectional traceability under the constraints of the bidirectional traceability authority;

[0035] The two-way traceability behavior authority includes a two-way traceability depth authority and a two-way traceability time authority. The two-way traceability depth authority is characterized by the traceability data query depth of the enterprise node that performs the two-way traceability behavior to the upstream enterprise node or the downstream enterprise node of the industrial chain. The two-way traceability time authority is characterized by the traceability data open time of the upstream enterprise node or the downstream enterprise node of the industrial chain to the enterprise node that performs the two-way traceability behavior, wherein,

[0036] The calculation formula for the two-way traceability depth authority is:

[0037]

[0038] Where high k It is represented by the bidirectional traceability depth of the k-th enterprise node, Represented as the traceability authority level of the k-th enterprise node k The upper limit of the two-way traceability depth, score k Represented as the traceability authority score of the k-th enterprise node, Represented as the traceability authority level of the k-th enterprise node k The lower limit of the confidence interval of the score;

[0039] The calculation formula for the two-way traceability time limit is:

[0040]

[0041] Where time k It is represented as the bidirectional traceability time of the k-th enterprise node. Represented as the traceability authority level of the k-th enterprise node k The upper limit of the two-way traceability time, score k Represented as the traceability authority score of the k-th enterprise node, Represented as the traceability authority level of the k-th enterprise node k The lower limit of the confidence interval of the score, k is a measurement constant.

[0042] As a preferred solution of the present invention, the bidirectional traceability under the bidirectional traceability behavior authority constraint includes:

[0043] The enterprise node executes a traceability data query to the upstream enterprise node or the downstream enterprise node of the industrial chain according to the traceability data query depth within the two-way traceability time limit to realize two-way traceability.

[0044] As a preferred solution of the present invention, the historical production behavior characteristic data and the historical traceability behavior characteristic data are normalized before being weighted and summed to obtain the traceability authority score of the enterprise node.

[0045] As a preferred embodiment of the present invention, the present invention provides a bidirectional traceability system according to the bidirectional traceability method of the steel industry chain, comprising:

[0046] A data processing unit is configured to sequentially acquire behavioral characteristic data representing the behavioral characteristics of enterprises at multiple enterprise nodes in the steel industry chain, mark the enterprise nodes with traceability authority levels based on the behavioral characteristic data, and map the behavioral characteristic data and traceability authority levels of the enterprise nodes into a traceability model sample sequence;

[0047] A model building unit is used to use a BP neural network to perform model training on the traceability model sample sequence to obtain a traceability model that identifies the traceability authority level of the enterprise node, so as to quantitatively map the traceability authority level to the enterprise behavior characteristics to achieve the centralization of traceability behavior based on the enterprise behavior characteristics;

[0048] The traceability execution unit is used to identify the traceability authority level of the enterprise nodes in the steel industry chain with the traceability authority level to be identified based on the traceability model, and control the enterprise nodes to perform two-way traceability in the steel industry chain according to the traceability authority level, so as to achieve the centralization of traceability behavior to reduce the retention of traceability data in the traceability behavior process.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] The present invention sequentially obtains behavioral feature data representing the behavioral features of enterprises at multiple enterprise nodes in the steel industry chain, marks the enterprise nodes with traceability authority levels based on the behavioral feature data, and establishes a traceability model so as to achieve centralization of traceability behavior based on enterprise behavioral features through quantitative mapping of traceability authority levels and enterprise behavioral features, and controls enterprise nodes to perform two-way traceability in the steel industry chain based on traceability authority levels, so as to reduce the amount of traceability data retained in the traceability behavior process based on centralization of traceability behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other implementation drawings based on the provided drawings without inventive effort.

[0052] Figure 1 A flow chart of a bidirectional traceability method provided by an embodiment of the present invention;

[0053] Figure 2 This is a structural block diagram of the bidirectional traceability system provided by an embodiment of the present invention.

[0054] The numbers in the figure represent the following:

[0055] 1-Data processing unit; 2-Model building unit; 3-Traceability execution unit. DETAILED DESCRIPTION

[0056] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0057] like Figure 1 As shown, the present invention provides a two-way traceability method for the steel industry chain, comprising the following steps:

[0058] Step S1: sequentially obtain behavioral feature data representing enterprise behavioral features at multiple enterprise nodes in the steel industry chain, mark the enterprise nodes with traceability authority levels based on the behavioral feature data, and map the behavioral feature data and traceability authority levels of the enterprise nodes into a traceability model sample sequence;

[0059] The enterprise behavior characteristics include historical production behavior characteristics and historical traceability behavior characteristics. The behavior characteristic data include historical production behavior characteristic data and historical traceability behavior characteristic data. The historical production behavior characteristics include but are not limited to historical output, historical production frequency, historical supply volume, and historical procurement volume, etc. The historical traceability behavior characteristic data include but are not limited to historical traceability behavior frequency, historical traceability behavior total volume, historical traceability complaint behavior frequency, and historical traceability complaint behavior total volume.

[0060] The tracing authority level marking of the enterprise node based on the behavioral feature data includes:

[0061] The traceability weights of the historical production behavior feature data and the historical traceability behavior feature data are set, and the historical production behavior feature data and the historical traceability behavior feature data are weighted and summed based on the traceability weights to obtain the traceability authority score of the enterprise node. The calculation formula of the traceability authority score is:

[0062] score i =α i X i +β i Y i ;

[0063] In the formula, score i Represented as the traceability authority score of the i-th enterprise node, X i、Y i Represented as the historical production behavior feature data and historical traceability behavior feature data of the i-th enterprise node, α i , β i They are respectively represented as the traceability weights of the historical production behavior feature data and the historical traceability behavior feature data of the i-th enterprise node;

[0064] The traceability authority score is matched with the score confidence interval of the traceability authority level, and the traceability authority level corresponding to the score confidence interval containing the traceability authority score is marked as the traceability authority level of the enterprise node. The higher the score, the higher the traceability authority level, which indicates that the traceability authority is greater. For example, the score confidence interval of the traceability authority level a is (A, B], and the score of an enterprise node is between A and B, which indicates that the traceability authority level of the enterprise node is a;

[0065] Preferably, the setting of the traceability weight includes:

[0066] The total data volume and total duration of the production behavior characteristic data and the historical tracing behavior characteristic data are counted respectively, and the average real-time data flow of the production behavior characteristic data and the historical tracing behavior characteristic data is calculated based on the total data volume and the total duration;

[0067] Normalizing the average real-time data stream of the production behavior characteristic data and the historical traceability behavior characteristic data to obtain the traceability weights of the production behavior characteristic data and the historical traceability behavior characteristic data;

[0068] The calculation formula of the traceability weight is:

[0069]

[0070] Where n i,X 、n i,Y The total amount of production behavior feature data and historical traceability behavior feature data of the i-th enterprise node, t i,X , t i,Y They are respectively represented as the total duration of the production behavior feature data and historical traceability behavior feature data of the i-th enterprise node, It is represented as the average real-time data stream of the production behavior feature data and historical traceability behavior feature data of the i-th enterprise node.

[0071] The larger the average real-time data flow of the production behavior characteristic data or the historical traceability behavior characteristic data is, the larger the data flow of the production behavior characteristic data or the historical traceability behavior characteristic data is mapped to the real scenario, which indicates that the enterprise node is an enterprise with high production volume and production frequency or an enterprise with large traceability demand. Regardless of which, it indicates that the enterprise node has a higher influence in the industrial chain. Therefore, the influence of the enterprise node is converted into a traceability authority score for quantification; and the larger the average real-time data flow of the production behavior characteristic data or the historical traceability behavior characteristic data is, the higher the possibility that the production behavior characteristic data or the historical traceability behavior characteristic data of the enterprise node contain important features. Therefore, a higher weight is given to the production behavior characteristic data or the historical traceability behavior characteristic data of the enterprise node to retain important feature data in the traceability authority score, making the calculation of the traceability authority score more credible.

[0072] Step S2: Using a BP neural network to train the traceability model sample sequence to obtain a traceability model that identifies the traceability authority level of enterprise nodes, so as to quantitatively map the traceability authority level to the enterprise behavior characteristics and realize the centralization of traceability behavior based on the enterprise behavior characteristics;

[0073] The mapping of the behavioral feature data and traceability authority level of the enterprise node into a traceability model sample sequence includes:

[0074] Map the behavior feature data and traceability authority level of the enterprise node into a sequence {X i ,Y i ,level i}, the sequence {X i ,Y i ,level i} as the sample sequence of the traceability model;

[0075] Among them, level i It represents the traceability authority level of the i-th enterprise node, i∈[1,m], m represents the total number of enterprise nodes, and i is a measurement constant.

[0076] The traceability model sample sequence is trained using a BP neural network to obtain a traceability model that identifies the traceability authority level of the enterprise node, including:

[0077] The traceability model sample sequence {X i ,Y i ,level i {X in} i ,Y i} as the input of BP neural network, and level iAs the output item of the BP neural network, the BP neural network is trained based on the input item and the output item to obtain a traceability model representing the mapping relationship between the behavior feature data and the traceability authority level;

[0078] The expression of the traceability model is:

[0079] level = BP(X,Y);

[0080] Among them, level represents the traceability authority level, {X, Y} represents the behavioral feature data, and BP represents the BP neural network.

[0081] Establishing a traceability model can directly calculate the traceability authority level of the enterprise node based on the behavioral characteristic data of the enterprise node, thereby improving computing efficiency.

[0082] Step S3: Based on the traceability model, the traceability authority level of the enterprise nodes to be identified in the steel industry chain is identified, and the enterprise nodes are controlled to perform two-way traceability in the steel industry chain according to the traceability authority level, so as to achieve centralization of traceability behavior to reduce the retention of traceability data in the traceability behavior process.

[0083] The method of identifying the traceability authority level of enterprise nodes in the steel industry chain with a traceability authority level to be identified based on the traceability model includes:

[0084] Extract the behavioral feature data of the enterprise node whose traceability authority level is to be identified in sequence {X j ,Y j}, and the behavioral characteristic data {X j ,Y j Input into the traceability model to obtain the traceability authority level of the enterprise node to be identified j , where X j 、Y j They are respectively represented as the production behavior feature data and historical traceability behavior feature data of the enterprise node of the jth traceability authority level to be identified, level j It is represented as the traceability authority level of the enterprise node with the jth traceability authority level to be identified.

[0085] The control of enterprise nodes in the steel industry chain based on traceability authority levels to perform two-way traceability includes:

[0086] Calculate the bidirectional traceability authority of the enterprise node based on the traceability authority level and traceability authority score of the enterprise node, and perform bidirectional traceability under the constraints of the bidirectional traceability authority;

[0087] The two-way traceability behavior authority includes a two-way traceability depth authority and a two-way traceability time authority. The two-way traceability depth authority is characterized by the traceability data query depth of the enterprise node that performs the two-way traceability behavior to the upstream enterprise node or the downstream enterprise node of the industrial chain. The two-way traceability time authority is characterized by the traceability data open time of the upstream enterprise node or the downstream enterprise node of the industrial chain to the enterprise node that performs the two-way traceability behavior, wherein,

[0088] The calculation formula for the two-way traceability depth authority is:

[0089]

[0090] Where high k It is represented by the bidirectional traceability depth of the k-th enterprise node, Represented as the traceability authority level of the k-th enterprise node k The upper limit of the two-way traceability depth, score k Represented as the traceability authority score of the k-th enterprise node, Represented as the traceability authority level of the k-th enterprise node k The lower limit of the confidence interval of the score;

[0091] The calculation formula for the two-way traceability time limit is:

[0092]

[0093] Where time k It is represented as the bidirectional traceability time of the k-th enterprise node. Represented as the traceability authority level of the k-th enterprise node k The upper limit of the two-way traceability time, score k Represented as the traceability authority score of the k-th enterprise node, Represented as the traceability authority level of the k-th enterprise node k The lower limit of the confidence interval of the score, k is a measurement constant.

[0094] By constraining permissions based on the depth and timeliness of two-way traceability, enterprise nodes with higher traceability authority levels, that is, enterprise nodes with higher influence in the industrial chain, can have higher two-way traceability depth and timeliness, which can solve the large number of traceability needs caused by large output and high production frequency in real scenarios. Moreover, for enterprise nodes with small output and slow production frequency in real scenarios, lower depth and timeliness can ensure that the traceability needs of enterprises are met while reducing the retention of the traceability data of the enterprise in the traceability system, and perform concurrency within a limited time limit to achieve controllable system retention and concurrency, thereby supplying the remaining resources to enterprise nodes with large enterprise traceability needs, realizing reasonable allocation of system resources, reducing data concurrency and data retention, and effectively alleviating system pressure and improving system efficiency.

[0095] The bidirectional traceability is performed under the bidirectional traceability behavior authority constraint, including:

[0096] The enterprise node executes a traceability data query to the upstream enterprise node or the downstream enterprise node of the industrial chain according to the traceability data query depth within the two-way traceability time limit to realize two-way traceability.

[0097] The historical production behavior characteristic data and the historical traceability behavior characteristic data are normalized before being weighted and summed to obtain the traceability authority score of the enterprise node.

[0098] like Figure 2 As shown, based on the above-mentioned two-way traceability method of the steel industry chain, the present invention provides a two-way traceability system, including:

[0099] Data processing unit 1 is used to sequentially obtain behavioral feature data representing enterprise behavioral features at multiple enterprise nodes in the steel industry chain, mark the enterprise nodes with traceability authority levels based on the behavioral feature data, and map the behavioral feature data and traceability authority levels of the enterprise nodes into a traceability model sample sequence;

[0100] Model building unit 2, used to use BP neural network to perform model training on the traceability model sample sequence to obtain a traceability model that identifies the traceability authority level of enterprise nodes, so as to quantitatively map the traceability authority level to the enterprise behavior characteristics and realize the centralization of traceability behavior based on the enterprise behavior characteristics;

[0101] The traceability execution unit 3 is used to identify the traceability authority level of the enterprise nodes in the steel industry chain whose traceability authority level is to be identified based on the traceability model, and control the enterprise nodes to perform two-way traceability in the steel industry chain according to the traceability authority level, so as to achieve the reduction of the traceability data retention in the traceability behavior process based on the centralization of the traceability behavior.

[0102] The present invention sequentially obtains behavioral feature data representing the behavioral features of enterprises at multiple enterprise nodes in the steel industry chain, marks the enterprise nodes with traceability authority levels based on the behavioral feature data, and establishes a traceability model so as to achieve centralization of traceability behavior based on enterprise behavioral features through quantitative mapping of traceability authority levels and enterprise behavioral features, and controls enterprise nodes to perform two-way traceability in the steel industry chain based on traceability authority levels, so as to reduce the amount of traceability data retained in the traceability behavior process based on centralization of traceability behavior.

[0103] The above embodiments are merely exemplary embodiments of the present application and are not intended to limit the scope of the present application. The scope of protection of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and scope of protection of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the scope of protection of the present application.

Claims

1. A two-way traceability method for the steel industry chain, characterized by: The following steps are involved: Step S1: sequentially obtain behavioral feature data representing enterprise behavioral features at multiple enterprise nodes in the steel industry chain, mark the enterprise nodes with traceability authority levels based on the behavioral feature data, and map the behavioral feature data and traceability authority levels of the enterprise nodes into a traceability model sample sequence; Step S2: Using a BP neural network to train the traceability model sample sequence to obtain a traceability model that identifies the traceability authority level of enterprise nodes, so as to quantitatively map the traceability authority level to the enterprise behavior characteristics and realize the centralization of traceability behavior based on the enterprise behavior characteristics; Step S3: Identify the traceability authority level of the enterprise nodes in the steel industry chain whose traceability authority level is to be identified based on the traceability model, and control the enterprise nodes to perform two-way traceability in the steel industry chain according to the traceability authority level, so as to reduce the amount of traceability data retained in the traceability process based on the centralization of traceability behavior; Control enterprise nodes in the steel industry chain to conduct two-way traceability based on traceability authority levels, including: The bidirectional traceability authority of the enterprise node is calculated based on the traceability authority level and traceability authority score of the enterprise node, and bidirectional traceability is performed under the constraints of the bidirectional traceability authority; The two-way traceability behavior authority includes the two-way traceability depth authority and the two-way traceability time authority. The two-way traceability depth authority is characterized by the traceability data query depth of the enterprise node that performs the two-way traceability behavior to the upstream enterprise node or the downstream enterprise node of the industrial chain. The two-way traceability time authority is characterized by the traceability data open time of the upstream enterprise node or the downstream enterprise node of the industrial chain to the enterprise node that performs the two-way traceability behavior. The calculation formula for the two-way traceability depth authority is: ； Where, It is represented by the bidirectional traceability depth of the k-th enterprise node, Represented as the traceability authority level of the k-th enterprise node The upper limit of the two-way traceability depth, Represented as the traceability authority score of the k-th enterprise node, Represented as the traceability authority level of the k-th enterprise node The lower limit of the confidence interval of the score; The calculation formula for the two-way traceability time limit is: ； Where, It is represented as the bidirectional traceability time of the k-th enterprise node. Represented as the traceability authority level of the k-th enterprise node The upper limit of the two-way traceability time limit, Represented as the traceability authority score of the k-th enterprise node, Represented as the traceability authority level of the k-th enterprise node The lower limit of the confidence interval of the score, k is a measurement constant.

2. The bidirectional traceability method for the steel industry chain according to claim 1, characterized in that: The enterprise behavior characteristics include historical production behavior characteristics and historical traceability behavior characteristics, and the behavior characteristic data include historical production behavior characteristic data and historical traceability behavior characteristic data.

3. The bidirectional traceability method for the steel industry chain according to claim 2, characterized in that: The tracing authority level marking of the enterprise node based on the behavioral feature data includes: The traceability weights of the historical production behavior feature data and the historical traceability behavior feature data are set, and the historical production behavior feature data and the historical traceability behavior feature data are weighted and summed based on the traceability weights to obtain the traceability authority score of the enterprise node. The calculation formula of the traceability authority score is: ； Where, Characterized as the i The traceability authority score of each enterprise node, 、 Characterized as the i The historical production behavior characteristic data and historical traceability behavior characteristic data of each enterprise node, 、 Respectively represented as i The traceability weight of the historical production behavior feature data and historical traceability behavior feature data of each enterprise node; Perform inclusion matching on the traceability authority score and the score confidence interval of the traceability authority level, and mark the traceability authority level corresponding to the score confidence interval containing the traceability authority score as the traceability authority level of the enterprise node; Preferably, the setting of the traceability weight includes: The total data volume and total duration of the historical production behavior characteristic data and the historical traceability behavior characteristic data are respectively counted, and the average real-time data flow of the historical production behavior characteristic data and the historical traceability behavior characteristic data is calculated based on the total data volume and the total duration; Normalizing the average real-time data stream of the historical production behavior characteristic data and the historical traceability behavior characteristic data to obtain the traceability weights of the historical production behavior characteristic data and the historical traceability behavior characteristic data; The calculation formula of the traceability weight is: ； ； Where, 、 Respectively represented as i The total amount of historical production behavior feature data and historical traceability behavior feature data of each enterprise node, 、 Respectively represented as i The total duration of the historical production behavior feature data and historical traceability behavior feature data of each enterprise node, 、 Characterized as the i The average real-time data stream of the historical production behavior characteristic data and historical traceability behavior characteristic data of each enterprise node.

4. The bidirectional traceability method for the steel industry chain according to claim 3, characterized in that: The mapping of the behavioral feature data and traceability authority level of the enterprise node into a traceability model sample sequence includes: Map the behavioral feature data and traceability authority level of the enterprise node into a sequence , convert the sequence As a sample sequence of the traceability model; in, Characterized as the i The traceability permission level of each enterprise node, , m represents the total number of enterprise nodes, i is a metrological constant.

5. The bidirectional traceability method for the steel industry chain according to claim 4, characterized in that: The traceability model sample sequence is trained using a BP neural network to obtain a traceability model that identifies the traceability authority level of an enterprise node, including: The traceability model sample sequence in As the input of the BP neural network, and As the output item of the BP neural network, the BP neural network is trained based on the input item and the output item to obtain a traceability model representing the mapping relationship between the behavior feature data and the traceability authority level; The expression of the traceability model is: ； in, Characterized by traceability authority level, Characterized as behavioral feature data, Characterized by BP neural network.

6. The bidirectional traceability method for the steel industry chain according to claim 5, characterized in that: The method of identifying the traceability authority level of enterprise nodes in the steel industry chain with a traceability authority level to be identified based on the traceability model includes: Extract the behavioral feature data of the enterprise nodes whose traceability authority levels are to be identified in sequence , and the behavioral characteristic data Input into the traceability model to obtain the traceability authority level of the enterprise node to be identified ,in, 、 Respectively represented as j The historical production behavior feature data and historical traceability behavior feature data of the enterprise node whose traceability authority level is to be identified, Characterized as the j The traceability permission level of the enterprise node to be identified.

7. A two-way traceability method for the steel industry chain according to claim 6, characterized in that: The bidirectional traceability is performed under the bidirectional traceability behavior authority constraint, including: The enterprise node executes a traceability data query to the upstream enterprise node or the downstream enterprise node of the industrial chain according to the traceability data query depth within the two-way traceability time limit to realize two-way traceability.

8. The bidirectional traceability method for the steel industry chain according to claim 3, characterized in that: The historical production behavior characteristic data and the historical traceability behavior characteristic data are normalized before being weighted and summed to obtain the traceability authority score of the enterprise node.

9. A bidirectional traceability system for the steel industry chain bidirectional traceability method according to any one of claims 1 to 8, characterized in that: include: A data processing unit (1) is used to sequentially obtain behavioral feature data representing the behavioral features of enterprises at multiple enterprise nodes in the steel industry chain, mark the enterprise nodes with traceability authority levels based on the behavioral feature data, and map the behavioral feature data and traceability authority levels of the enterprise nodes into a traceability model sample sequence; A model building unit (2) is used to train the traceability model sample sequence using a BP neural network to obtain a traceability model that identifies the traceability authority level of the enterprise node, so as to achieve the centralization of the traceability behavior based on the enterprise behavior characteristics by quantitatively mapping the traceability authority level with the enterprise behavior characteristics; The traceability execution unit (3) is used to identify the traceability authority level of the enterprise node to be identified in the steel industry chain based on the traceability model, and control the enterprise node to perform two-way traceability in the steel industry chain according to the traceability authority level, so as to reduce the retention amount of traceability data in the traceability process based on the centralization of traceability behavior.

Images