A pipe full life cycle RFID intelligent traceability management method and system and medium

Abstract

The application discloses a kind of pipe full life cycle RFID intelligent traceability management method, system and medium, it is related to life cycle management field, including the following steps: obtaining the RFID reading data of target pipe, pipe operation and maintenance data and service environment data;Pipe state transition unit set is constructed;According to pipe state transition unit set, pipe operation and maintenance data and service environment data, generate state feature set;Perform feature discretization, and generate the life cycle state entropy corresponding to target pipe;According to life cycle state entropy, generate the stability determination result corresponding to target pipe, and perform state grade back, generate state entropy source representation;Generate the service risk determination result of target pipe;Determine the inspection priority, maintenance disposal type and traceability query path corresponding to target pipe.The present application realizes pipe full life cycle risk traceability using RFID state entropy, with the advantages of timely identification and accurate disposal.

Description

Technical Field

[0001] This invention relates to the field of lifecycle management, and in particular to a method, system and medium for intelligent RFID traceability management of pipe materials throughout their entire lifecycle. Background Technology

[0002] Currently, RFID tags are typically used to record the identity information of pipes throughout their lifecycle, including production, quality inspection, warehousing, transportation, installation, inspection, maintenance, and scrapping. Reading and writing devices are used to query the location of pipes, node status, and operation and maintenance records to achieve pipe source tracking, node record verification, and basic file management.

[0003] However, existing RFID traceability management for pipes mostly focuses on identification and record querying. It is difficult to transform scattered RFID reading data, pipe operation and maintenance data, and service environment data into quantitative results that characterize the stability of pipe status. It is also difficult to identify the source of status entropy and generate service risk assessment results, making it difficult to determine the inspection priority, maintenance and disposal type, and traceability query path in a timely and accurate manner. Summary of the Invention

[0004] One objective of this invention is to propose a method, system, and medium for intelligent traceability management of pipe materials throughout their entire lifecycle using RFID. This invention utilizes RFID state entropy to achieve risk traceability of pipe materials throughout their entire lifecycle, and has the advantages of timely identification and accurate handling.

[0005] A method for intelligent traceability management of pipe materials throughout their entire lifecycle according to an embodiment of the present invention includes the following steps: Acquire RFID reading data, pipe operation and maintenance data, and service environment data of the target pipe at the stages of production, quality inspection, warehousing, transportation, installation, inspection, maintenance, and scrapping. Based on RFID reading data, a state transition unit for the target pipe is constructed between adjacent life cycle nodes, generating a set of pipe state transition units; Based on the set of pipe state transition units, pipe operation and maintenance data, and service environment data, extract the number of state transitions, number of abnormal nodes, maintenance frequency, location change magnitude, number of changes in operating entities, and intensity of service environment changes corresponding to the target pipe to generate a set of state features; Perform feature discretization on the set of state features, determine the state distribution corresponding to each state feature, and generate the life cycle state entropy corresponding to the target pipe based on the state distribution; Based on the life cycle state entropy, the stability judgment result corresponding to the target pipe is generated, and the state level is reversed item by item according to the discrete state level corresponding to each state feature in the state feature set to generate the state entropy source representation. Based on the stability assessment results and the characterization of the source of state entropy, the service risk assessment results of the target pipe are generated; Based on the service risk assessment results, the inspection priority, maintenance and treatment type and traceability query path corresponding to the target pipe are determined, and intelligent traceability management results are generated.

[0006] Optionally, the RFID reading data includes pipe identification code, RFID reading time, RFID reading location and life cycle node identifier; pipe operation and maintenance data includes anomaly type, anomaly location, maintenance frequency and maintenance result; and service environment data includes burial depth, medium pressure, ambient humidity, corrosion level and external load level.

[0007] Optionally, the construction of the pipe state transition unit set specifically includes: Based on the pipe identification code in the RFID reading data, the reading records formed at each life cycle node of the same target pipe are collected; The collected reading records are sorted according to the RFID reading time, and the order of production, quality inspection, warehousing, transportation, installation, inspection, maintenance and scrapping nodes corresponding to the reading records is determined according to the life cycle node identifier. Two read records corresponding to adjacent sorting positions are identified as a node transfer pair. The preceding node identifier, following node identifier, preceding read time, following read time, preceding read position, and following read position in the node transfer pair are read. Based on the preceding node identifier, the following node identifier, the preceding read time, the following read time, the preceding read position, and the following read position, a state transition unit for the target pipe between adjacent life cycle nodes is constructed. Arrange the state transition units sequentially to generate a set of pipe state transition units.

[0008] Optionally, the generation of the state feature set specifically includes: The number of state transition units in the pipe state transition unit set is counted to obtain the number of state jumps. The preceding and following read positions in each state transition unit are read to determine the position change range. The number of times the operating entity changes between adjacent lifecycle nodes is counted to obtain the number of changes in the operating entity; Based on the anomaly type, location, number of maintenance sessions, and maintenance results in the pipe maintenance data, determine the number of anomaly nodes and the maintenance frequency; The intensity of service environment changes is determined based on the burial depth, medium pressure, ambient humidity, corrosion level, and external load level in the service environment data. The number of state transitions, the number of abnormal nodes, the frequency of maintenance, the magnitude of location changes, the number of changes in the operating entity, and the intensity of changes in the service environment are collected to generate a set of state characteristics.

[0009] Optionally, the generation of the lifecycle state entropy specifically includes: Read the number of state transitions, the number of abnormal nodes, the maintenance frequency, the magnitude of location changes, the number of changes in the operating entity, and the intensity of changes in the service environment from the status feature set; Normalization is performed on the number of state transitions, the number of abnormal nodes, the maintenance frequency, the magnitude of location changes, the number of changes in the operating subject, and the intensity of changes in the service environment, respectively, to generate a set of normalized state features; Discretize each normalized state feature in the normalized state feature set to determine the discrete state level corresponding to each normalized state feature; The number of times each discrete state level appears in the normalized state feature set is counted, and the state distribution corresponding to the target pipe is determined based on the proportion of the number of occurrences to the total number of normalized state features. Based on the proportion of each discrete state level in the state distribution, the degree of dispersion of the state distribution is determined, and the degree of dispersion of the state distribution is used as the life cycle state entropy of the target pipe.

[0010] Optionally, the generation of the state entropy source representation specifically includes: Obtain the historical life cycle state entropy of the same type of pipes during their historical life cycle process, and determine the state entropy benchmark value by sorting the historical life cycle state entropy from smallest to largest. The life cycle state entropy corresponding to the target pipe is compared with the state entropy benchmark value. When the life cycle state entropy is not greater than the state entropy benchmark value, the stability judgment result corresponding to the stable state is generated. When the lifecycle state entropy is greater than the state entropy baseline value, a stability determination result corresponding to the unstable state is generated. Read the discrete state levels corresponding to the number of state transitions, number of abnormal nodes, maintenance frequency, location change magnitude, number of changes in operating subjects, and intensity of changes in service environment in the state feature set, and determine each state feature as a state feature to be back-inferred. For any state feature to be reversed, the discrete state level corresponding to the state feature to be reversed is replaced with the first discrete state level, while keeping the discrete state levels corresponding to other state features unchanged. The occurrence frequency and proportion of each discrete state level are recounted to generate the reversed state distribution, and the reversed life cycle state entropy is generated based on the reversed state distribution. Based on the difference between the life cycle state entropy and each inverse life cycle state entropy, the state entropy reduction contribution value corresponding to each state feature is determined, and the state feature with the largest state entropy reduction contribution value is determined as the state entropy source feature, and a state entropy source representation is generated.

[0011] Optionally, the generation of the service risk assessment result specifically includes: Read the stability determination results and the characterization of the source of state entropy; When the stability determination result is a stable state, the risk level of the target pipe is determined to be low risk, and the name of the state entropy source feature in the corresponding state entropy source characterization is used as the risk source feature. When the stability determination result is an unstable state, read the name of the state entropy source feature, the corresponding discrete state level, and the state entropy reduction contribution value from the state entropy source representation. Based on the name of the source feature of state entropy, the risk source feature of the target pipe is determined, and the candidate results of risk level are determined based on the corresponding discrete state level. The state entropy reduction contribution value is compared with the preset contribution threshold, and the risk level candidate results are adjusted according to the comparison results to obtain the risk level corresponding to the target pipe. By associating risk level with risk source characteristics, the service risk assessment result corresponding to the target pipe material is generated.

[0012] Optionally, the generation of the intelligent traceability management result specifically includes: Read the risk level and risk source characteristics from the service risk assessment results; Inspection priorities are determined based on risk levels. When the risk level is low, medium-low, medium-high, or high, it is determined as the fourth, third, second, or first inspection priority, respectively. The maintenance and handling type is determined based on the characteristics of the risk source. When the characteristics of the risk source are the number of status jumps, the magnitude of location changes, or the number of changes in the operating entity, the maintenance and handling type is determined to be the verification of the transfer record. When the risk source characteristics are the number of abnormal nodes or the frequency of maintenance, the maintenance handling type is determined to be operation and maintenance review and handling. When the risk source is characterized by the intensity of changes in the service environment, the maintenance and treatment type is determined to be service environment retesting; Determine the source tracing path based on the characteristics of the risk source; By associating inspection priorities, maintenance and disposal types, and traceability query paths, intelligent traceability management results are generated.

[0013] According to an embodiment of the present invention, a pipe material full lifecycle RFID intelligent traceability management system includes: The data acquisition module is used to acquire RFID reading data, pipe operation and maintenance data, and service environment data of the target pipe at the nodes of production, quality inspection, warehousing, transportation, installation, inspection, maintenance, and scrapping. The state transition construction module is used to construct state transition units between adjacent life cycle nodes of the target pipe based on RFID reading data, and generate a set of pipe state transition units; The status feature extraction module is used to extract the number of status jumps, number of abnormal nodes, maintenance frequency, location change magnitude, number of changes in operating subjects, and intensity of changes in the service environment corresponding to the target pipe based on the pipe status transition unit set, pipe operation and maintenance data, and service environment data, and generate a status feature set. The state entropy generation module is used to perform feature discretization on the set of state features, determine the state distribution corresponding to each state feature, and generate the life cycle state entropy corresponding to the target pipe based on the state distribution. The state entropy source analysis module is used to generate the stability judgment result corresponding to the target pipe based on the life cycle state entropy, and to perform state level back-inference item by item based on the discrete state level corresponding to each state feature in the state feature set to generate the state entropy source characterization. The service risk assessment module is used to generate service risk assessment results for the target pipe based on the stability assessment results and the characterization of the state entropy source. The traceability management module is used to determine the inspection priority, maintenance and treatment type and traceability query path of the target pipe material based on the service risk assessment results, and generate intelligent traceability management results.

[0014] According to an embodiment of the present invention, a computer-readable storage medium stores a computer program that, when executed by a processor, enables the processor to perform a pipe material full lifecycle RFID intelligent traceability management method.

[0015] The beneficial effects of this invention are: This invention acquires RFID reading data, pipe operation and maintenance data, and service environment data corresponding to the production, quality inspection, warehousing, transportation, installation, inspection, maintenance, and scrapping stages of target pipes. This enables the unified utilization of scattered identification records, operation and maintenance records, and service environment records throughout the entire lifecycle of the pipe. Compared to methods relying solely on RFID tags for pipe identification or node tracing, this invention further constructs state transition units between adjacent lifecycle nodes based on RFID reading data. This ensures that the flow of the target pipe between different lifecycle nodes has a clear temporal sequence, node boundaries, and positional correspondence, providing a continuous and traceable data foundation for subsequent state feature extraction.

[0016] This invention, based on a set of pipe state transition units, pipe operation and maintenance data, and service environment data, extracts the number of state transitions, the number of abnormal nodes, maintenance frequency, the magnitude of location changes, the number of changes in operating entities, and the intensity of service environment changes. It can characterize the state changes of the target pipe from multiple dimensions, including node flow, abnormal distribution, maintenance intervention, spatial changes, changes in operating entities, and fluctuations in the service environment. By performing feature discretization on the state feature set and generating a state distribution, and then generating a lifecycle state entropy based on the state distribution, it can transform state change information from different sources and with different dimensions into a unified quantification result of state stability. This avoids the shortcomings of existing technologies that rely solely on single record readings or human experience to judge pipe state stability.

[0017] This invention further generates stability assessment results based on lifecycle state entropy, and generates a state entropy source characterization by performing a back-calculation of the state level for each discrete state characteristic. This identifies the main state characteristics that have a significant impact on lifecycle state entropy, ensuring that service risk assessment not only provides risk level results but also corresponding risk source evidence. Based on this, the invention determines inspection priorities, maintenance and handling types, and traceability query paths according to the service risk assessment results. This enables pipe inspection, maintenance, and traceability to shift from post-event queries to proactive management based on state changes, improving the timeliness of hidden service risk identification, the targeting of maintenance and handling, and the accuracy of traceability queries. Attached Figure Description

[0018] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of an RFID intelligent traceability management method for the entire life cycle of pipes proposed in this invention; Figure 2 This is a comparison diagram of the state distribution and the reversed state distribution of the RFID intelligent traceability management method for the entire life cycle of pipes proposed in this invention. Detailed Implementation

[0019] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0020] refer to Figure 1 and Figure 2 A method for intelligent traceability management of pipe materials throughout their entire lifecycle using RFID, comprising the following steps: Acquire RFID reading data, pipe operation and maintenance data, and service environment data of the target pipe at the stages of production, quality inspection, warehousing, transportation, installation, inspection, maintenance, and scrapping. Based on RFID reading data, a state transition unit for the target pipe is constructed between adjacent life cycle nodes, generating a set of pipe state transition units; Based on the set of pipe state transition units, pipe operation and maintenance data, and service environment data, extract the number of state transitions, number of abnormal nodes, maintenance frequency, location change magnitude, number of changes in operating entities, and intensity of service environment changes corresponding to the target pipe to generate a set of state features; Perform feature discretization on the set of state features, determine the state distribution corresponding to each state feature, and generate the life cycle state entropy corresponding to the target pipe based on the state distribution; Based on the life cycle state entropy, the stability judgment result corresponding to the target pipe is generated, and the state level is reversed item by item according to the discrete state level corresponding to each state feature in the state feature set to generate the state entropy source representation. Based on the stability assessment results and the characterization of the source of state entropy, the service risk assessment results of the target pipe are generated; Based on the service risk assessment results, the inspection priority, maintenance and treatment type and traceability query path corresponding to the target pipe are determined, and intelligent traceability management results are generated.

[0021] In this embodiment, RFID reading data is an identification record formed by reading the RFID tags of the target pipe at each life cycle node using an RFID reader / writer device. This includes the pipe identification code, RFID reading time, RFID reading location, and life cycle node identifier. Pipe operation and maintenance data is the operation and maintenance record formed by the target pipe at the inspection, repair, and scrapping nodes. This includes the anomaly type, anomaly location, maintenance frequency, and maintenance results. Service environment data is the external usage environment record corresponding to the target pipe at the installation, inspection, repair, and scrapping nodes. This includes the burial depth, medium pressure, ambient humidity, corrosion level, and external load level.

[0022] In this embodiment, the construction of the pipe state transition unit set specifically includes: Based on the pipe identification code in the RFID reading data, the reading records formed at each life cycle node of the same target pipe are collected; Pipe identification code is the coding information written into the RFID tag of the target pipe and used to uniquely identify the target pipe. It is used to associate the RFID reading data generated by the same target pipe at each life cycle node. The collected reading records are sorted according to the RFID reading time, and the order of production, quality inspection, warehousing, transportation, installation, inspection, maintenance and scrapping nodes corresponding to the reading records is determined according to the life cycle node identifier. Lifecycle node identifiers include production node identifiers, quality inspection node identifiers, warehousing node identifiers, transportation node identifiers, installation node identifiers, patrol inspection node identifiers, maintenance node identifiers, and scrapping node identifiers; Two read records corresponding to adjacent sorting positions are identified as a node transfer pair. The preceding node identifier, following node identifier, preceding read time, following read time, preceding read position, and following read position in the node transfer pair are read. Based on the preceding node identifier, the following node identifier, the preceding read time, the following read time, the preceding read position, and the following read position, a state transition unit for the target pipe between adjacent life cycle nodes is constructed. Arrange the state transition units according to the life cycle nodes of the target pipe to generate a set of pipe state transition units; This invention can collect RFID reading records scattered across different lifecycle nodes of a target pipe according to the pipe's identification code, and organize them sequentially based on RFID reading time and lifecycle node identifiers. This transforms the flow process between adjacent lifecycle nodes into state transition units with preceding node identifiers, subsequent node identifiers, preceding reading time, subsequent reading time, preceding reading position, and subsequent reading position. In this way, scattered RFID reading records are constructed into a continuous set of pipe state transition units, providing a clear data foundation for extracting the number of state jumps and the magnitude of position changes. Furthermore, it ensures that the state change process upon which the calculation of lifecycle state entropy depends has a clear temporal sequence, node boundaries, and positional correspondence.

[0023] In this embodiment, the generation of the state feature set specifically includes: The number of state transition units in the pipe state transition unit set is counted to obtain the number of state jumps. The preceding and following read positions in each state transition unit are read to determine the position change range. In practice, each state transition unit in the pipe state transition unit set is read, each state transition unit is counted as one state jump, and the number of state jumps is accumulated for each state transition unit; the preceding and following read positions in each state transition unit are read, the position difference between the preceding and following read positions is calculated, the position difference value corresponding to each state transition unit is obtained, and the position difference values ​​are accumulated to obtain the position change range of the target pipe during its life cycle; The number of times the operating entity changes between adjacent lifecycle nodes is counted to obtain the number of changes in the operating entity; In practice, the operation subject identifier is obtained, and the preceding and subsequent operation subject identifiers corresponding to adjacent life cycle nodes are determined according to the pipe status transition unit set. The preceding and subsequent operation subject identifiers corresponding to the same status transition unit are compared. When the two are different, the status transition unit is counted as one operation subject change. The number of operation subject changes corresponding to each status transition unit is accumulated to obtain the number of operation subject changes of the target pipe during its life cycle. The operating entities are the personnel, work teams, equipment, or management units that actually perform operations at the stages of production, quality inspection, warehousing, transportation, installation, patrol inspection, maintenance, or scrapping of the target pipe materials; Based on the anomaly type, location, number of maintenance sessions, and maintenance results in the pipe maintenance data, determine the number of anomaly nodes and the maintenance frequency; In practice, the system reads the anomaly type, anomaly location, maintenance frequency, and maintenance result from the pipe maintenance data. Based on the anomaly type and location, it identifies the lifecycle nodes with anomaly records and counts the number of lifecycle nodes with anomaly records to obtain the number of anomaly nodes. Based on the maintenance frequency, it reads the number of maintenance records formed at the inspection, repair, and scrapping nodes of the target pipe and, combined with the maintenance results, removes unexecuted or invalid maintenance records to count the number of valid maintenance records to obtain the maintenance frequency. The intensity of service environment changes is determined based on the burial depth, medium pressure, ambient humidity, corrosion level, and external load level in the service environment data. In practice, the burial depth, medium pressure, ambient humidity, corrosion level, and external load level are read from the service environment data according to the life cycle nodes of the target pipe. For adjacent acquisition nodes, the changes in burial depth, medium pressure, ambient humidity, corrosion level, and external load level are calculated respectively, and each change value is normalized to obtain the normalized change values ​​for depth, pressure, humidity, corrosion, and external load. The five normalized change values ​​are weighted according to preset environmental weights to obtain the node environmental change value corresponding to the adjacent acquisition node. The node environmental change values ​​corresponding to each adjacent acquisition node are accumulated to obtain the service environment change intensity corresponding to the target pipe. The preset environmental weights include a weight of 0.15 for burial depth, 0.25 for medium pressure, 0.15 for ambient humidity, 0.25 for corrosion level, and 0.20 for external load level. Among them, medium pressure, corrosion level, and external load level have a greater impact on the service stability of the pipe and have higher weights than burial depth and ambient humidity. Burial depth is the underground burial distance of the target pipe relative to the ground surface after installation; medium pressure is the pressure value of the medium transported inside the target pipe during service; ambient humidity is the humidity level of the surrounding soil, pipe gallery, or installation environment of the target pipe; corrosion level is the classification of the level of corrosion impact of the external environment on the pipe; external load level is the classification of the level of stress impact of vehicle rolling, soil pressure, construction disturbance, or other external forces on the target pipe. Burial depth, medium pressure, ambient humidity, corrosion level, and external load level are all environmental factors that can affect the stability of the target pipe in its service environment. Among them, burial depth reflects the degree of influence of soil cover and construction environment on the pipe; medium pressure reflects the degree of pressure change inside the pipe; ambient humidity and corrosion level reflect the degree of change of external corrosion conditions; and external load level reflects the degree of change of external forces such as vehicle rolling, soil pressure, or construction disturbance. Since the above data have different dimensions, value ranges, and degrees of influence on the stability of pipe service, each change value is first normalized and then weighted according to the preset environmental weights. This can unify different types of environmental factors into comparable environmental change characterization values, thereby obtaining the intensity of service environment change corresponding to the target pipe more accurately. The number of status transitions is used to characterize the frequency of the flow changes of the target pipe between life cycle nodes such as production, quality inspection, warehousing, transportation, installation, inspection, maintenance and scrapping. The number of abnormal nodes is used to characterize the distribution range of nodes with abnormal records during the life cycle of the target pipe. The maintenance frequency is used to characterize the frequency of maintenance activities such as inspection and maintenance of the target pipe during operation and maintenance. The position change range is used to characterize the overall span of the spatial position change of the target pipe between different life cycle nodes. The number of changes in the operating subject is used to characterize the degree of change of personnel, teams, equipment or management units involved in the operation of the target pipe during the life cycle flow. The intensity of service environment change is used to characterize the comprehensive degree of change of environmental factors such as burial depth, medium pressure, ambient humidity, corrosion level and external load level of the target pipe during installation and service. The greater the number of state transitions, the more frequent the node transitions; the greater the number of abnormal nodes, the more abnormal nodes are covered; the greater the maintenance frequency, the more frequent the operation and maintenance interventions; the greater the location change range, the greater the spatial transition span; the greater the number of changes in the operating entity, the more frequent the changes in the operational responsibility chain; and the greater the intensity of changes in the service environment, the more obvious the fluctuations in the external service environment. The number of state transitions, the number of abnormal nodes, the maintenance frequency, the magnitude of location changes, the number of changes in the operating entity, and the intensity of changes in the service environment are collected to generate a set of state characteristics. This invention, through collaborative analysis of pipe state transition unit sets, node operation data, pipe maintenance data, and service environment data, can extract state transition counts, abnormal node counts, maintenance frequency, location change magnitude, number of operator changes, and service environment change intensity from multiple dimensions such as life cycle flow, anomaly distribution, maintenance intervention, spatial changes, operator changes, and service environment changes. This allows the state changes of the target pipe throughout its entire life cycle to no longer rely solely on a single RFID reading record for judgment, but instead form a set of state features that can uniformly characterize the stability of the pipe's state.

[0024] In this embodiment, the generation of lifecycle state entropy specifically includes: Read the number of state transitions, the number of abnormal nodes, the maintenance frequency, the magnitude of location changes, the number of changes in the operating entity, and the intensity of changes in the service environment from the status feature set; Normalization is performed on the number of state transitions, the number of abnormal nodes, the maintenance frequency, the magnitude of location changes, the number of changes in the operating subject, and the intensity of changes in the service environment, respectively, to generate a set of normalized state features; Discretize each normalized state feature in the normalized state feature set to determine the discrete state level corresponding to each normalized state feature; In specific implementation, the normalized values ​​of state transition, abnormal node, maintenance frequency, location change, operation subject change, and service environment change are read from the normalized state feature set; the normalized value range is divided into the first discrete interval [0, 0.25), the second discrete interval [0.25, 0.50), the third discrete interval [0.50, 0.75), and the fourth discrete interval [0.75, 1.00]; the normalized state features falling into the first discrete interval are determined as the first discrete state level, the normalized state features falling into the second discrete interval are determined as the second discrete state level, the normalized state features falling into the third discrete interval are determined as the third discrete state level, and the normalized state features falling into the fourth discrete interval are determined as the fourth discrete state level, thereby determining the discrete state level corresponding to each normalized state feature; The number of times each discrete state level appears in the normalized state feature set is counted, and the state distribution corresponding to the target pipe is determined based on the proportion of the number of occurrences to the total number of normalized state features. In practice, the discrete state levels corresponding to the normalized values ​​of state transitions, abnormal nodes, maintenance frequency, location changes, operating entity changes, and service environment changes are read. The number of times each of the first, second, third, and fourth discrete state levels appears in the above six normalized state features is counted to obtain the number of times the first, second, third, and fourth levels appear. The number of times each discrete state level appears is divided by the total number of normalized state features (six) to obtain the proportion value corresponding to each discrete state level. The proportion values ​​of the first, second, third, and fourth levels together constitute the state distribution corresponding to the target pipe. The specific form of the state distribution is written as a discrete level ratio sequence, which consists of the first discrete state level ratio value, the second discrete state level ratio value, the third discrete state level ratio value, and the fourth discrete state level ratio value, expressed as: State distribution = [first level ratio value, second level ratio value, third level ratio value, fourth level ratio value]; where each ratio value is equal to the number of occurrences of the corresponding discrete state level divided by the total number of normalized state features, and the sum of the four ratio values ​​is 1, which is used to characterize the distribution of the six state features of the target pipe at different state change levels; Based on the proportion of each discrete state level in the state distribution, the degree of dispersion of the state distribution is determined, and the degree of dispersion of the state distribution is used as the life cycle state entropy of the target pipe. In practice, the proportions of the first, second, third, and fourth levels in the state distribution are read. For discrete state levels with a proportion of 0, the discrete contribution value corresponding to that discrete state level is set to 0. For discrete state levels with a proportion greater than 0, a natural logarithmic transformation is performed on the proportion, and the product of the logarithmic transformation result and the proportion is multiplied and the negative number is taken to obtain the discrete contribution value corresponding to that discrete state level. The discrete contribution values ​​of the first, second, third, and fourth levels are accumulated to obtain the degree of dispersion of the state distribution. In this embodiment, the lifecycle state entropy is used to characterize the degree of dispersion of the six state characteristics of the target pipe at different discrete state levels, rather than characterizing the numerical value of a single state characteristic. When multiple state characteristics fall concentratedly into the same discrete state level, it indicates that the source of state changes in the target pipe is relatively concentrated, and the lifecycle state entropy is low. When multiple state characteristics are dispersed into different discrete state levels, it indicates that the target pipe exhibits differentiated fluctuations in multiple aspects such as circulation, anomalies, maintenance, location, operating entity, and service environment, and the lifecycle state entropy is high. In this way, the dispersion and uncertainty of the state changes of the target pipe throughout its entire lifecycle can be identified, providing a basis for subsequent stability determination and characterization of the source of state entropy.

[0025] In this embodiment, the generation of the state entropy source representation specifically includes: Obtain the historical life cycle state entropy of the same type of pipes during their historical life cycle process, and determine the state entropy benchmark value by sorting the historical life cycle state entropy from smallest to largest. In practice, the historical life cycle state entropy located at the 75th percentile after sorting is selected as the state entropy benchmark value; when the 75th percentile is not an integer sorting position, the corresponding historical life cycle state entropy is read according to the sorting position after rounding up, and it is determined as the state entropy benchmark value. The life cycle state entropy corresponding to the target pipe is compared with the state entropy benchmark value. When the life cycle state entropy is not greater than the state entropy benchmark value, the stability judgment result corresponding to the stable state is generated. When the lifecycle state entropy is greater than the state entropy baseline value, a stability determination result corresponding to the unstable state is generated. Read the discrete state levels corresponding to the number of state transitions, number of abnormal nodes, maintenance frequency, location change magnitude, number of changes in operating subjects, and intensity of changes in service environment in the state feature set, and determine each state feature as a state feature to be back-inferred. The state features to be reversed are those that are selected one by one and whose discrete state levels are temporarily adjusted during the analysis of the source of state entropy, in order to determine their degree of influence on the life cycle state entropy. For any state feature to be reversed, the discrete state level corresponding to the state feature to be reversed is replaced with the first discrete state level, while keeping the discrete state levels corresponding to other state features unchanged. The occurrence frequency and proportion of each discrete state level are recounted to generate the reversed state distribution, and the reversed life cycle state entropy is generated based on the reversed state distribution. This invention replaces the discrete state level corresponding to the state feature to be reversed with the first discrete state level, while keeping the discrete state levels corresponding to other state features unchanged. The purpose is to simulate the change in state entropy after the state feature recovers from its current change level to the lowest change level, thereby observing the degree of influence of this state feature on the lifecycle state entropy. If the decrease in the reversed lifecycle state entropy after replacement is significant, it indicates that the state feature has a strong influence on the dispersion of the target pipe's state distribution. In this way, the entropy reduction contribution of each state feature can be identified one by one without changing other state features, avoiding the need to judge the source of state entropy solely based on the original numerical value, making the characterization of the source of state entropy more accurate and traceable. In practice, the reverse-engineered lifecycle state entropy is generated one by one according to the state characteristics. Each time, only one state characteristic is selected from the following: number of state transitions, number of abnormal nodes, maintenance frequency, location change magnitude, number of changes in operating entities, and intensity of service environment changes. The discrete state level corresponding to this selected state characteristic is then replaced with the first discrete state level, while the discrete state levels corresponding to other state characteristics remain unchanged. The resulting reverse-engineered state distribution and reverse-engineered lifecycle state entropy only reflect the impact of a single state characteristic recovering to its lowest change level on the lifecycle state entropy. Through six reverse-engineering iterations, the reverse-engineered lifecycle state entropy corresponding to each of the six state characteristics can be obtained, and the state entropy reduction contribution value corresponding to each state characteristic can be calculated accordingly. Based on the difference between the life cycle state entropy and each inverse life cycle state entropy, the state entropy reduction contribution value corresponding to each state feature is determined, and the state feature with the largest state entropy reduction contribution value is determined as the state entropy source feature, and a state entropy source representation is generated. In practice, the current lifecycle state entropy of the target pipe is read, and the number of state transitions, the number of abnormal nodes, the maintenance frequency, the magnitude of position changes, the number of changes in the operating entity, and the intensity of changes in the service environment are read as the back-calculated lifecycle state entropy generated when back-calculating the state features. For any state feature, the difference between the current lifecycle state entropy and the back-calculated lifecycle state entropy corresponding to that state feature is calculated. When the current lifecycle state entropy is greater than the back-calculated lifecycle state entropy, the difference between the two is determined as the state entropy reduction contribution value corresponding to that state feature. When the current lifecycle state entropy is not greater than the back-calculated lifecycle state entropy, the state entropy reduction contribution value corresponding to that state feature is determined as 0. The difference calculation of the six state features is completed in sequence to obtain the state entropy reduction contribution value corresponding to each state feature. The state entropy source characterization includes the state entropy source feature name, the corresponding discrete state level, and the state entropy reduction contribution value; wherein, the state entropy source feature name is used to identify the main state feature that causes the target pipe's life cycle state entropy to increase, the corresponding discrete state level is used to characterize the current degree of change of the state feature, and the state entropy reduction contribution value is used to characterize the strength of the influence of the state feature on the life cycle state entropy. The source feature of state entropy is the state feature that contributes the most to the reduction of state entropy in the back-analysis.

[0026] In this embodiment, the generation of service risk assessment results specifically includes: Read the stability determination results and the characterization of the source of state entropy; When the stability determination result is a stable state, the risk level of the target pipe is determined to be low risk, and the name of the state entropy source feature in the corresponding state entropy source characterization is used as the risk source feature. The representation of the source of state entropy in a steady state is only used to record the state features that have the greatest impact on the life cycle state entropy in the current set of state features, and is not used as a basis for upgrading the risk level. When the stability determination result is an unstable state, read the name of the state entropy source feature, the corresponding discrete state level, and the state entropy reduction contribution value from the state entropy source representation. Based on the name of the source feature of state entropy, the risk source feature of the target pipe is determined, and the candidate results of risk level are determined based on the corresponding discrete state level. In practice, the corresponding discrete state level in the state entropy source representation is read, and a correspondence is established between the discrete state level and the candidate risk level results. When the corresponding discrete state level is the first discrete state level, the candidate risk level result is determined as a low-risk candidate result; when the corresponding discrete state level is the second discrete state level, the candidate risk level result is determined as a medium-low-risk candidate result; when the corresponding discrete state level is the third discrete state level, the candidate risk level result is determined as a medium-high-risk candidate result; when the corresponding discrete state level is the fourth discrete state level, the candidate risk level result is determined as a high-risk candidate result, thus obtaining the candidate risk level result corresponding to the target pipe material. The state entropy reduction contribution value is compared with the preset contribution threshold, and the risk level candidate results are adjusted according to the comparison results to obtain the risk level corresponding to the target pipe. In practice, the state entropy reduction contribution value in the state entropy source representation is read and compared with a preset contribution threshold. When the state entropy reduction contribution value is less than the preset contribution threshold, the risk level candidate result determined according to the corresponding discrete state level remains unchanged, and this risk level candidate result is determined as the risk level corresponding to the target pipe. When the state entropy reduction contribution value is greater than or equal to the preset contribution threshold, the risk level candidate result is adjusted up one level to obtain the risk level corresponding to the target pipe. Among them, the low-risk candidate result is adjusted to medium-low risk, the medium-low-risk candidate result is adjusted to medium-high risk, the medium-high-risk candidate result is adjusted to high risk, and the high-risk candidate result remains high risk. In practice, the historical state entropy reduction contribution value corresponding to each state characteristic in the historical samples of the same type of pipe is obtained, and the historical state entropy reduction contribution value is sorted in ascending order of value. The historical state entropy reduction contribution value corresponding to the 75th percentile position is selected as the preset contribution threshold. In this embodiment, the preset contribution threshold is 0.08. By associating risk level with risk source characteristics, the service risk assessment result corresponding to the target pipe material is generated.

[0027] In this embodiment, the generation of intelligent traceability management results specifically includes: Read the risk level and risk source characteristics from the service risk assessment results; Inspection priorities are determined based on risk levels. When the risk level is low, medium-low, medium-high, or high, it is determined as the fourth, third, second, or first inspection priority, respectively. The maintenance and handling type is determined based on the characteristics of the risk source. When the characteristics of the risk source are the number of status jumps, the magnitude of location changes, or the number of changes in the operating entity, the maintenance and handling type is determined to be the verification of the transfer record. When the risk source characteristics are the number of abnormal nodes or the frequency of maintenance, the maintenance handling type is determined to be operation and maintenance review and handling. When the risk source is characterized by the intensity of changes in the service environment, the maintenance and treatment type is determined to be service environment retesting; The flow record verification addresses the source of state entropy caused by the number of state jumps, the magnitude of location changes, or the number of changes in the operating entity. It reviews the RFID reading data, pipe state transition unit set, and node operation data corresponding to the target pipe, checking for any missed readings, misreadings, abnormal node sequences, abnormal location records, or abnormal changes in the operating entity, in accordance with the pipe identification code, life cycle node identifier, RFID reading time, RFID reading location, and operating entity identifier. The maintenance review addresses the source of state entropy caused by the number of abnormal nodes or maintenance frequency. It reviews the abnormality type, abnormality location, maintenance frequency, and maintenance results in the pipe maintenance data, and conducts on-site verification of the corresponding inspection nodes, maintenance nodes, or scrapped nodes. The service environment retest addresses the source of state entropy caused by the intensity of changes in the service environment. It re-collects and compares the burial depth, medium pressure, ambient humidity, corrosion level, and external load level corresponding to the target pipe to confirm whether the changes in the service environment persist. Determine the source tracing path based on the characteristics of the risk source; In practice, the risk source characteristics are read, and the corresponding traceability query path is determined based on these characteristics. When the risk source characteristics are the number of status jumps, the magnitude of location changes, or the number of changes in the operating entity, the RFID reading data, pipe status transfer unit sets, and node operation data corresponding to the production node, quality inspection node, warehousing node, transportation node, and installation node are determined as the traceability query path. When the risk source characteristics are the number of abnormal nodes or the maintenance frequency, the RFID reading data and pipe operation and maintenance data corresponding to the inspection node, maintenance node, and scrapping node are determined as the traceability query path. When the risk source characteristics are the intensity of changes in the service environment, the RFID reading data and service environment data corresponding to the installation node, inspection node, maintenance node, and scrapping node are determined as the traceability query path. By associating inspection priorities, maintenance and disposal types, and traceability query paths, intelligent traceability management results are generated.

[0028] A pipe material full lifecycle RFID intelligent traceability management system, comprising: The data acquisition module is used to acquire RFID reading data, pipe operation and maintenance data, and service environment data of the target pipe at the nodes of production, quality inspection, warehousing, transportation, installation, inspection, maintenance, and scrapping. The state transition construction module is used to construct state transition units between adjacent life cycle nodes of the target pipe based on RFID reading data, and generate a set of pipe state transition units; The status feature extraction module is used to extract the number of status jumps, number of abnormal nodes, maintenance frequency, location change magnitude, number of changes in operating subjects, and intensity of changes in the service environment corresponding to the target pipe based on the pipe status transition unit set, pipe operation and maintenance data, and service environment data, and generate a status feature set. The state entropy generation module is used to perform feature discretization on the set of state features, determine the state distribution corresponding to each state feature, and generate the life cycle state entropy corresponding to the target pipe based on the state distribution. The state entropy source analysis module is used to generate the stability judgment result corresponding to the target pipe based on the life cycle state entropy, and to perform state level back-inference item by item based on the discrete state level corresponding to each state feature in the state feature set to generate the state entropy source characterization. The service risk assessment module is used to generate service risk assessment results for the target pipe based on the stability assessment results and the characterization of the state entropy source. The traceability management module is used to determine the inspection priority, maintenance and treatment type and traceability query path of the target pipe material based on the service risk assessment results, and generate intelligent traceability management results.

[0029] A computer-readable storage medium storing a computer program that, when executed by a processor, enables the processor to perform a pipe material full lifecycle RFID intelligent traceability management method.

[0030] Example 1: To verify the feasibility of this invention in practice, it was applied to the operation and maintenance site of a gas pipeline network in a new urban area. Approximately 3.6 kilometers of new medium-pressure gas pipelines were laid in this area, with a total of 428 PE gas pipes under management. These pipes undergo multiple lifecycle stages before and after being put into use, including production, quality inspection, warehousing, transportation, installation, inspection, and maintenance. The existing management method relied primarily on RFID tags to query pipe identification and transfer records, which could confirm the origin and passage of a particular pipe. However, when pipes underwent multiple transfers, installation location adjustments, frequent changes in operating entities, an increase in abnormal inspection records, or changes in the service environment, management personnel needed to meticulously review all records and maintenance logs. This made it difficult to promptly determine whether the pipes presented any hidden service risks, and also made it difficult to quickly identify the pipes to be inspected first, the types of maintenance to be implemented, and the tracing points to be investigated.

[0031] In this scenario, each target pipe has a unique pipe identification code written into its RFID tag. At production, quality inspection, warehousing, transportation, installation, inspection, maintenance, and scrapping nodes, RFID readers read the pipe identification code, RFID reading time, RFID reading location, and lifecycle node identifier to form RFID reading data. Simultaneously, the system collects anomaly types, anomaly locations, maintenance frequency, and maintenance results at inspection, maintenance, and scrapping nodes to form pipe operation and maintenance data. It also collects the burial depth, medium pressure, ambient humidity, corrosion level, and external load level corresponding to installation, inspection, maintenance, and scrapping nodes to form service environment data. The system aggregates the RFID reading records of the same target pipe at each lifecycle node according to the pipe identification code, then determines the node order according to the RFID reading time and lifecycle node identifier, constructing state transition units between adjacent lifecycle nodes, transforming the originally scattered reading records into a continuous set of pipe state transition units.

[0032] Taking the target pipe with the serial number GC-A-217 as an example, this pipe formed 6 effective state transition units during the management cycle, with a cumulative position change range of 0.92 kilometers and 4 registered changes in the operating entity. During the inspection and maintenance process, one life cycle node with an abnormal record was identified, with 1 effective maintenance record. After installation, the burial depth of this pipe changed from 1.25 meters to 1.29 meters, the medium pressure fluctuated between 0.36 MPa and 0.41 MPa, the ambient humidity increased from 62% to 66%, the corrosion level remained at level 1, and the external load level increased from level 2 to level 3. The service environment change intensity was calculated to be 0.23 according to the preset environmental weight. Based on this, the system obtained the number of state transitions, the number of abnormal nodes, the maintenance frequency, the position change range, the number of changes in the operating entity, and the service environment change intensity, and aggregated the six types of state characteristics into a state characteristic set.

[0033] The system normalizes the six state features mentioned above, resulting in normalized state features of GC-A-217 of 0.18, 0.22, 0.19, 0.21, 0.80, and 0.23, respectively. After discretizing the features according to the first discrete interval [0, 0.25), the second discrete interval [0.25, 0.50), the third discrete interval [0.50, 0.75), and the fourth discrete interval [0.75, 1.00], the number of state transitions, the number of abnormal nodes, the maintenance frequency, the magnitude of location changes, and the intensity of service environment changes all fall into the first discrete state level, while the number of changes in the operating entity falls into the fourth discrete state level. Thus, the state distribution is [5 / 6, 0, 0, 1 / 6]. Based on the proportions corresponding to each discrete state level in the state distribution, the degree of dispersion of the state distribution is calculated, resulting in a lifecycle state entropy of approximately 0.451 for the target pipe. The 75th percentile of the historical life cycle state entropy of similar pipes on site is 0.36. Therefore, the life cycle state entropy corresponding to GC-A-217 is greater than the state entropy benchmark value, and the system generates the stability judgment result corresponding to the unstable state.

[0034] Subsequently, the system performs state level back-calculation item by item based on the discrete state level corresponding to each state feature in the state feature set. For the number of state transitions, the number of abnormal nodes, maintenance frequency, location change magnitude, and service environment change intensity, since they are currently at the first discrete state level, replacing them with the first discrete state level does not reduce the back-calculated lifecycle state entropy, and the corresponding state entropy reduction contribution value is 0. For the number of changes in the operating subject, the system replaces its corresponding fourth discrete state level with the first discrete state level, while keeping the discrete state levels corresponding to other state features unchanged, and re-counts the back-calculated state distribution as [1, 0, 0, 0], with a back-calculated lifecycle state entropy of 0 and a state entropy reduction contribution value of 0.451. Since the state entropy reduction contribution value corresponding to the number of changes in the operating subject is the largest, the system determines the number of changes in the operating subject as the state entropy source feature and generates a state entropy source representation including the state entropy source feature name, the corresponding discrete state level, and the state entropy reduction contribution value.

[0035] Based on the stability assessment results and the characterization of state entropy sources, the system identifies the risk source characteristic of GC-A-217 as the number of changes in the operating entity, and determines the candidate risk level as a high-risk candidate based on its corresponding fourth discrete state level. Since the state entropy reduction contribution value of 0.451 is greater than the preset contribution threshold of 0.08, and the high-risk candidate is already at the highest level, the risk level of the target pipe material is ultimately maintained as high risk. The system further assigns high risk to the first inspection priority, determines the maintenance and handling type corresponding to the risk source characteristic "number of changes in the operating entity" as flow record verification, and determines the RFID reading data, pipe material state transfer unit set, and operating entity identifiers registered at each lifecycle node corresponding to the production node, quality inspection node, warehousing node, transportation node, and installation node as the traceability query path.

[0036] During 90 consecutive days of field application, the system completed batch analysis of 428 pipes, identifying 9 high-risk pipes and 29 medium-to-high-risk pipes, and prioritizing high-risk pipes such as GC-A-217 to the inspection list. On-site verification confirmed that the high risk of GC-A-217 was indeed concentrated in the process of frequent changes in the operating entity. This was because the pipe was registered and operated by four different work teams during warehousing, transfer, and installation handover, with some handover records being supplemented. Compared to the original manual method of checking each pipe individually, the risk location time for a single pipe was reduced from approximately 18 minutes to approximately 5.1 minutes, the accuracy rate of risk source location increased from 76.4% to 92.0%, and the number of invalid inspections decreased by approximately 36.5%. This demonstrates that the present invention can transform scattered RFID reading data, pipe maintenance data, and service environment data into quantitative results that can be used for stability assessment and risk tracing, improving the timeliness of identifying hidden service risks in pipes, the targeting of maintenance and handling, and the accuracy of tracing queries.

[0037] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for intelligent traceability management of pipe materials throughout their entire lifecycle using RFID, characterized in that, Includes the following steps: Acquire RFID reading data, pipe operation and maintenance data, and service environment data of the target pipe at the stages of production, quality inspection, warehousing, transportation, installation, inspection, maintenance, and scrapping. Based on RFID reading data, a state transition unit for the target pipe is constructed between adjacent life cycle nodes, generating a set of pipe state transition units; Based on the set of pipe state transition units, pipe operation and maintenance data, and service environment data, extract the number of state transitions, number of abnormal nodes, maintenance frequency, location change magnitude, number of changes in operating entities, and intensity of service environment changes corresponding to the target pipe to generate a set of state features; Perform feature discretization on the set of state features, determine the state distribution corresponding to each state feature, and generate the life cycle state entropy corresponding to the target pipe based on the state distribution; Based on the life cycle state entropy, the stability judgment result corresponding to the target pipe is generated, and the state level is reversed item by item according to the discrete state level corresponding to each state feature in the state feature set to generate the state entropy source representation. Based on the stability assessment results and the characterization of the source of state entropy, the service risk assessment results of the target pipe are generated; Based on the service risk assessment results, the inspection priority, maintenance and treatment type and traceability query path corresponding to the target pipe are determined, and intelligent traceability management results are generated.

2. The RFID intelligent traceability management method for the entire life cycle of pipe materials according to claim 1, characterized in that, The RFID reading data includes pipe identification code, RFID reading time, RFID reading location and life cycle node identifier. The pipe operation and maintenance data includes anomaly type, anomaly location, maintenance frequency and maintenance result. The service environment data includes burial depth, medium pressure, ambient humidity, corrosion level and external load level.

3. The RFID intelligent traceability management method for the entire life cycle of pipe materials according to claim 1, characterized in that, The construction of the pipe state transition unit set specifically includes: Based on the pipe identification code in the RFID reading data, the reading records formed at each life cycle node of the same target pipe are collected; The collected reading records are sorted according to the RFID reading time, and the order of production, quality inspection, warehousing, transportation, installation, inspection, maintenance and scrapping nodes corresponding to the reading records is determined according to the life cycle node identifier. Two read records corresponding to adjacent sorting positions are identified as a node transfer pair. The preceding node identifier, following node identifier, preceding read time, following read time, preceding read position, and following read position in the node transfer pair are read. Based on the preceding node identifier, the following node identifier, the preceding read time, the following read time, the preceding read position, and the following read position, a state transition unit for the target pipe between adjacent life cycle nodes is constructed. Arrange the state transition units sequentially to generate a set of pipe state transition units.

4. The RFID intelligent traceability management method for the entire life cycle of pipe materials according to claim 1, characterized in that, The generation of the state feature set specifically includes: The number of state transition units in the pipe state transition unit set is counted to obtain the number of state jumps. The preceding and following read positions in each state transition unit are read to determine the position change range. The number of times the operating entity changes between adjacent lifecycle nodes is counted to obtain the number of changes in the operating entity; Based on the anomaly type, location, number of maintenance sessions, and maintenance results in the pipe maintenance data, determine the number of anomaly nodes and the maintenance frequency; The intensity of service environment changes is determined based on the burial depth, medium pressure, ambient humidity, corrosion level, and external load level in the service environment data. The number of state transitions, the number of abnormal nodes, the frequency of maintenance, the magnitude of location changes, the number of changes in the operating entity, and the intensity of changes in the service environment are collected to generate a set of state characteristics.

5. The RFID intelligent traceability management method for the entire life cycle of pipe materials according to claim 1, characterized in that, The generation of the lifecycle state entropy specifically includes: Read the number of state transitions, the number of abnormal nodes, the maintenance frequency, the magnitude of location changes, the number of changes in the operating entity, and the intensity of changes in the service environment from the status feature set; Normalization is performed on the number of state transitions, the number of abnormal nodes, the maintenance frequency, the magnitude of location changes, the number of changes in the operating subject, and the intensity of changes in the service environment, respectively, to generate a set of normalized state features; Discretize each normalized state feature in the normalized state feature set to determine the discrete state level corresponding to each normalized state feature; The number of times each discrete state level appears in the normalized state feature set is counted, and the state distribution corresponding to the target pipe is determined based on the proportion of the number of occurrences to the total number of normalized state features. Based on the proportion of each discrete state level in the state distribution, the degree of dispersion of the state distribution is determined, and the degree of dispersion of the state distribution is used as the life cycle state entropy of the target pipe.

6. The method for intelligent traceability management of pipe materials throughout their entire lifecycle using RFID, as described in claim 1, is characterized in that... The generation of the state entropy source representation specifically includes: Obtain the historical life cycle state entropy of the same type of pipes during their historical life cycle process, and determine the state entropy benchmark value by sorting the historical life cycle state entropy from smallest to largest. The life cycle state entropy corresponding to the target pipe is compared with the state entropy benchmark value. When the life cycle state entropy is not greater than the state entropy benchmark value, the stability judgment result corresponding to the stable state is generated. When the lifecycle state entropy is greater than the state entropy baseline value, a stability determination result corresponding to the unstable state is generated. Read the discrete state levels corresponding to the number of state transitions, number of abnormal nodes, maintenance frequency, location change magnitude, number of changes in operating subjects, and intensity of changes in service environment in the state feature set, and determine each state feature as a state feature to be back-inferred. For any state feature to be reversed, the discrete state level corresponding to the state feature to be reversed is replaced with the first discrete state level, while keeping the discrete state levels corresponding to other state features unchanged. The occurrence frequency and proportion of each discrete state level are recounted to generate the reversed state distribution, and the reversed life cycle state entropy is generated based on the reversed state distribution. Based on the difference between the life cycle state entropy and each inverse life cycle state entropy, the state entropy reduction contribution value corresponding to each state feature is determined, and the state feature with the largest state entropy reduction contribution value is determined as the state entropy source feature, and a state entropy source representation is generated.

7. The RFID intelligent traceability management method for the entire life cycle of pipe materials according to claim 1, characterized in that, The generation of the service risk assessment result specifically includes: Read the stability determination results and the characterization of the source of state entropy; When the stability determination result is a stable state, the risk level of the target pipe is determined to be low risk, and the name of the state entropy source feature in the corresponding state entropy source characterization is used as the risk source feature. When the stability determination result is an unstable state, read the name of the state entropy source feature, the corresponding discrete state level, and the state entropy reduction contribution value from the state entropy source representation. Based on the name of the source feature of state entropy, the risk source feature of the target pipe is determined, and the candidate results of risk level are determined based on the corresponding discrete state level. The state entropy reduction contribution value is compared with the preset contribution threshold, and the risk level candidate results are adjusted according to the comparison results to obtain the risk level corresponding to the target pipe. By associating risk level with risk source characteristics, the service risk assessment result corresponding to the target pipe material is generated.

8. The RFID intelligent traceability management method for the entire life cycle of pipe materials according to claim 1, characterized in that, The generation of the intelligent traceability management results specifically includes: Read the risk level and risk source characteristics from the service risk assessment results; Inspection priorities are determined based on risk levels. When the risk level is low, medium-low, medium-high, or high, it is determined as the fourth, third, second, or first inspection priority, respectively. The maintenance and handling type is determined based on the characteristics of the risk source. When the characteristics of the risk source are the number of status jumps, the magnitude of location changes, or the number of changes in the operating entity, the maintenance and handling type is determined to be the verification of the transfer record. When the risk source characteristics are the number of abnormal nodes or the frequency of maintenance, the maintenance handling type is determined to be operation and maintenance review and handling. When the risk source is characterized by the intensity of changes in the service environment, the maintenance and treatment type is determined to be service environment retesting; Determine the source tracing path based on the characteristics of the risk source; By associating inspection priorities, maintenance and disposal types, and traceability query paths, intelligent traceability management results are generated.

9. A pipe material full lifecycle RFID intelligent traceability management system, implementing the pipe material full lifecycle RFID intelligent traceability management method according to any one of claims 1 to 8, characterized in that, include: The data acquisition module is used to acquire RFID reading data, pipe operation and maintenance data, and service environment data of the target pipe at the nodes of production, quality inspection, warehousing, transportation, installation, inspection, maintenance, and scrapping. The state transition construction module is used to construct state transition units between adjacent life cycle nodes of the target pipe based on RFID reading data, and generate a set of pipe state transition units; The status feature extraction module is used to extract the number of status jumps, number of abnormal nodes, maintenance frequency, location change magnitude, number of changes in operating subjects, and intensity of changes in the service environment corresponding to the target pipe based on the pipe status transition unit set, pipe operation and maintenance data, and service environment data, and generate a status feature set. The state entropy generation module is used to perform feature discretization on the set of state features, determine the state distribution corresponding to each state feature, and generate the life cycle state entropy corresponding to the target pipe based on the state distribution. The state entropy source analysis module is used to generate the stability judgment result corresponding to the target pipe based on the life cycle state entropy, and to perform state level back-inference item by item based on the discrete state level corresponding to each state feature in the state feature set to generate the state entropy source characterization. The service risk assessment module is used to generate service risk assessment results for the target pipe based on the stability assessment results and the characterization of the state entropy source. The traceability management module is used to determine the inspection priority, maintenance and treatment type and traceability query path of the target pipe material based on the service risk assessment results, and generate intelligent traceability management results.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables the processor to perform the RFID intelligent traceability management method for the entire lifecycle of pipes as described in any one of claims 1 to 8.

Images