Pipeline full-life-cycle management system based on MA identifier and cryptographic algorithm

Abstract

The invention discloses a pipeline full life cycle management system based on MA identification and a national secret algorithm, and relates to the technical field of pipeline management, a data acquisition and security processing module is constructed to acquire pipeline static information of each pipeline, and a unique identifier is marked and encrypted; the digital asset ledger building module takes the unique identifier as an index and builds a digital asset ledger of the pipeline; constructing a life cycle management module to track the state of each stage of the pipeline from production to scrapping to generate full life cycle data of the pipeline; an intelligent analysis module is constructed to predict the state of the pipeline based on the full-life-cycle data of the pipeline, and an analysis report of the pipeline is generated; according to the method, the credible data base is constructed by fusing MA unified identification and a national secret algorithm, and a multi-agent collaborative optimization mechanism based on digital portraits is introduced, so that intelligent management and control of the whole life cycle of the pipeline are realized.

Description

Technical Field

[0001] This invention relates to the field of pipeline management technology, specifically to a pipeline lifecycle management system based on MA identifiers and national cryptographic algorithms. Background Technology

[0002] Currently, the full lifecycle management of oil drilling tubing mainly relies on a combination of several decentralized technical solutions; asset identification often uses internal enterprise coding or simple steel stamps, data management relies on a combination of basic databases and paper records, and business functions are developed around independent modules for warehousing, well running records, and maintenance registration. However, existing technologies have certain limitations. First, the lack of a unified identification system makes it difficult to identify pipelines when they are transferred between different units, creating traceability gaps. Furthermore, the use of general encryption algorithms cannot meet the national cryptographic management requirements for core industrial facilities, posing a risk of data leakage and tampering. At the same time, the decentralized system architecture and weak integration capabilities create information silos between production, operation, and maintenance, resulting in outdated and inaccurate asset ledger data and a fragmented view of the entire life cycle. Finally, existing technologies rely entirely on human experience for pipeline scheduling, maintenance, and scrapping decisions, lacking data-driven life prediction, risk warning, and optimized configuration capabilities. This leads to low asset management efficiency and potential safety risks in downhole operations caused by overuse or improper configuration. Therefore, it is of great significance to develop a pipeline lifecycle management system based on MA identification and national cryptographic algorithms. Summary of the Invention

[0003] The purpose of this invention is to provide a pipeline lifecycle management system based on MA identifiers and national cryptographic algorithms to address the shortcomings in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a pipeline lifecycle management system based on MA identifiers and national cryptographic algorithms, comprising: Data acquisition and security processing module: Collects static pipeline information for each pipeline and marks it with a unique identifier. Based on the national commercial cryptographic algorithm, it securely encrypts the unique identifier and its associated static pipeline information. Digital Asset Ledger Module: Connected to the data acquisition and security processing module, it uses unique identifiers as indexes to build a digital asset ledger for the pipeline; Lifecycle Management Module: Connects to the Digital Asset Ledger Module to track the status of pipelines at each stage from production to disposal, generate pipeline status tracking records, and transmit the status tracking records to the Digital Asset Ledger for updating the Digital Asset Ledger and generating full lifecycle data of the pipelines; Intelligent Analysis Module: Connects to the lifecycle management module, and based on the pipeline's full lifecycle data, predicts the pipeline's status and generates pipeline analysis reports.

[0005] In a preferred embodiment, the data acquisition and security processing module includes: The identification encoding and writing unit is used to obtain the pipeline MA unified identification code and its static information from the associated system, and write the MA unified identification code and its static information as a unique identifier to the physical carrier attached to the pipeline. The data encryption and decryption unit is used to encrypt and decrypt unique identifiers by calling national commercial cryptographic algorithms during data transmission and storage. The integrity verification unit is used to verify the integrity of pipeline static information based on national commercial cryptographic algorithms during data reading.

[0006] In a preferred embodiment, the digital asset ledger module includes: The information input and verification unit is used to receive the static information of the pipeline and to perform standardization and compliance verification on the static information of the pipeline. The ledger dynamic maintenance unit is used to link the static information of pipelines with real-time monitoring data using a unique identifier as an index, and to build a digital asset ledger. The comprehensive query unit, based on the digital asset ledger, allows users to perform combined queries based on multiple conditions and generate query results through multi-dimensional statistical analysis. The query results can also be visualized and exported.

[0007] In a preferred embodiment, the lifecycle management module includes: The lifecycle start-of-life recording unit creates a complete asset entry for the corresponding pipeline in the digital asset ledger based on the pipeline static information obtained after decrypting the unique identifier. The process recording unit is used to record the deployment location and operation parameters of the pipeline as real-time monitoring data based on the unique identifier identified during the pipeline's commissioning process, and to update the real-time monitoring data to the digital asset ledger. The status transition unit is used to update the status identifier of the pipeline in the digital asset ledger based on the inspection results after the pipeline recycling inspection. The scrapping process management unit is used to execute an online approval process for pipelines whose status is marked as pending scrapping, and after the approval is completed, update their status to scrapped in the digital asset ledger and attach an asset freeze mark. The pipeline lifecycle management module outputs full lifecycle data of the pipeline, which includes static information, real-time monitoring data and status indicators.

[0008] In a preferred embodiment, the intelligent analysis module includes: The digital profile building unit acquires pipeline lifecycle data and builds a dynamically updated multi-dimensional feature vector for each pipeline based on the pipeline lifecycle data, which serves as the digital profile of the pipeline. The pipeline intelligent agent modeling unit, based on digital profiling, creates a pipeline intelligent agent model with autonomous decision-making logic for each pipeline. The pipeline intelligent agent model includes the digital profiling of the pipeline and the environmental adaptability evaluation function. The virtual environment interaction unit is used to construct a virtual interaction environment that simulates the target downhole working conditions, and to put multiple candidate pipeline intelligent agent models into the virtual interaction environment for collaborative operation simulation. The group collaborative optimization unit is used to simulate and evaluate the interaction of different pipeline combinations in a virtual environment and generate the optimal pipeline or pipeline combination scheme. The optimal piping or piping combination scheme will be presented in the analysis report.

[0009] In a preferred embodiment, the step of creating a pipeline intelligent agent model with autonomous decision-making logic for each pipeline based on digital profiling, wherein the pipeline intelligent agent model includes a digital profiling of the pipeline and an environmental adaptability evaluation function, is as follows: Based on the pipeline's entire lifecycle data, extract the pipeline's static information, real-time monitoring data, and status indicators; Based on feature extraction algorithms, the pipeline static information, real-time monitoring data and status indicators are normalized and feature encoded to generate multi-dimensional feature vectors as digital profiles. The digital profile is used as the initial model for the pipeline agent model, and behavioral policies are applied to the initial model to generate the pipeline agent model. Among them, the behavioral strategies include deformation under environmental pressure loads, stress response thresholds, and coordination rules with other pipeline agents in load distribution. The training sample set is formed by using real-time monitoring data and its corresponding environmental parameters as model input features and the performance degradation index derived from the state identifier sequence as training target label. The training sample set is input into the neural network model to generate the environmental fitness evaluation function of the pipeline agent model.

[0010] In a preferred embodiment, the step of constructing a virtual interactive environment simulating the target downhole working condition and placing multiple candidate pipeline intelligent agent models into the environment for collaborative operation simulation is as follows: Based on the design requirements of the target downhole conditions, key environmental parameters are extracted, including well depth, formation pressure, temperature gradient, and concentration of corrosive components in the medium. Key environmental parameters are normalized to generate the initial input conditions for the virtual interactive environment. Input the initial input conditions into the preset physical rule model to generate a virtual interactive environment; Multiple pipeline agent models are transmitted to a virtual interactive environment to generate virtual interactive environment instances. The virtual interactive environment instance is driven to run according to the preset operation sequence, and the behavior strategies and environmental adaptability evaluation functions of all pipeline intelligent agent models are activated synchronously to simulate collaborative operations.

[0011] In a preferred embodiment, the step of simulating and evaluating the interaction of different pipe combinations in a virtual environment to generate the optimal pipe or pipe combination scheme is as follows: In a virtual interactive environment, the pipeline intelligent agent model that drives the input performs multi-round collaborative operation simulations based on behavioral strategies and environmental fitness evaluation functions; Based on the real-time performance of multi-round collaborative operation simulation, an overall performance index is generated to evaluate different combinations. A target optimization algorithm is used to iteratively adjust the combination of pipeline agent models; When the overall performance index reaches its optimal level, the corresponding pipeline intelligent agent model is combined, and the output is the optimal pipeline or pipeline combination scheme.

[0012] The technical effects and advantages provided by the present invention in the above technical solution are as follows: 1. This invention establishes a dual trust mechanism based on MA unified identification code and national cryptographic algorithm, assigning each pipeline an unalterable and unique identity, and performing end-to-end security encryption and integrity verification on its static and dynamic data throughout its entire life cycle. This solution addresses to some extent the long-standing pain points in the petroleum industry, such as chaotic asset identification, numerous data silos, and insufficient security compliance in information transmission and storage. It constructs a trusted data closed loop that runs through all stages of production, well deployment, recovery, and disposal, laying a solid data foundation for efficient asset traceability and refined management. 2. This invention transforms pipeline lifecycle data into dynamic digital profiles and uses these profiles as the core to construct a pipeline intelligent agent model with autonomous decision-making capabilities. Multi-agent collaborative operation simulation and optimization are then performed in a virtual environment simulating downhole working conditions. This solution overcomes the limitations of traditional asset allocation based on human experience and static rules, realizing a rule-based transformation in pipeline asset management from passive recording to proactive prediction, and from single-pipe management to group system optimization. Through simulation-driven optimization, it outputs the globally optimal pipeline combination scheme, significantly improving asset utilization, extending service life, and reducing downhole operation risks. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0014] Figure 1 This is a system flowchart of the present invention.

[0015] Figure 2 This is a logic block diagram of the present invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Example 1, please refer to Figure 1 and Figure 2 As shown in this embodiment, the pipeline lifecycle management system based on MA identifiers and national cryptographic algorithms... Data acquisition and security processing module: Collects static pipeline information for each pipeline and marks it with a unique identifier. Based on the national commercial cryptographic algorithm, it securely encrypts the unique identifier and its associated static pipeline information. Digital Asset Ledger Module: Connected to the data acquisition and security processing module, it uses unique identifiers as indexes to build a digital asset ledger for the pipeline; Lifecycle Management Module: Connects to the Digital Asset Ledger Module to track the status of pipelines at each stage from production to disposal, generate pipeline status tracking records, and transmit the status tracking records to the Digital Asset Ledger for updating the Digital Asset Ledger and generating full lifecycle data of the pipelines; Intelligent Analysis Module: Connects to the Lifecycle Management Module, and based on the pipeline's entire lifecycle data, predicts the pipeline's status and generates pipeline analysis reports; Furthermore, currently, the full lifecycle management of oil drilling tubing mainly relies on a combination of several decentralized technical solutions; asset identification often uses internal enterprise coding or simple steel stamps, data management relies on a combination of basic databases and paper records, while business functions revolve around independent storage, well run-in records, and maintenance registration modules. However, existing technologies have certain limitations. First, the lack of a unified identification system makes it difficult to identify pipelines when they are transferred between different units, creating traceability gaps. Furthermore, the use of general encryption algorithms cannot meet the national cryptographic management requirements for core industrial facilities, posing a risk of data leakage and tampering. At the same time, the decentralized system architecture and weak integration capabilities create information silos between production, operation, and maintenance, resulting in outdated and inaccurate asset ledger data and a fragmented view of the entire life cycle. Finally, existing technologies rely entirely on human experience for pipeline scheduling, maintenance, and scrapping decisions, lacking data-driven life prediction, risk warning, and optimized configuration capabilities. This leads to low asset management efficiency and potential safety risks in downhole operations caused by overuse or improper configuration. This invention establishes a dual trust mechanism based on the MA unified identifier code and national cryptographic algorithms, assigning each pipeline an unalterable and unique identity, and performing end-to-end security encryption and integrity verification on its static and dynamic data throughout its entire lifecycle. This solution addresses, to some extent, the long-standing pain points in the petroleum industry, such as chaotic asset identification, numerous data silos, and insufficient security compliance in information transmission and storage. It constructs a trusted data closed loop that runs through all stages of production, well deployment, recovery, and disposal, laying a solid data foundation for efficient asset traceability and refined management. Meanwhile, by transforming pipeline lifecycle data into dynamic digital profiles and using these as the core to build a pipeline intelligent agent model with autonomous decision-making capabilities, multi-agent collaborative operation simulation and optimization are carried out in a virtual environment simulating downhole working conditions. This solution breaks through the limitations of traditional asset allocation based on human experience and static rules, realizing the rule transformation of pipeline asset management from passive recording to active prediction, and from single-pipe management to group system optimization. Through simulation-driven optimization, it outputs the pipeline combination scheme with the best global performance, significantly improving asset utilization, extending service life, and reducing downhole operation risks.

[0018] In one embodiment, the data acquisition and security processing module includes: The identification encoding and writing unit is used to obtain the pipeline MA unified identification code and its static information from the associated system, and write the MA unified identification code and its static information as a unique identifier to the physical carrier attached to the pipeline. The data encryption and decryption unit is used to encrypt and decrypt unique identifiers by calling national commercial cryptographic algorithms during data transmission and storage. The integrity verification unit is used to verify the integrity of pipeline static information based on national commercial cryptographic algorithms during data reading; Furthermore, the identification encoding and writing unit automatically obtains the pipeline's unified identification code (MA) and associated static pipeline information assigned by an authoritative institution by calling the standard application programming interface (API) of the production enterprise resource planning system. This static information includes the pipeline's production information, material properties, dimensions, initial test data, and quality and certification information. Then, it drives an RFID reader connected to an industrial computer via a universal serial communication interface to write the complete data packet containing the MA unified identification code and pipeline static information into a passive UHF RFID tag fixed at the pipeline coupling, according to a standard data frame format. The data encryption and decryption unit pre-installs commercial cryptographic algorithms approved by the State Cryptography Administration in the system server and well site terminal equipment. In the Faku system, when a unique identifier needs to be transmitted over the network or stored in a database, the data encryption unit automatically calls the SM4 block cipher algorithm and uses an encryption key dynamically generated by the system key management service to encrypt the data and generate ciphertext. When the data needs to be read, the corresponding key is used to decrypt it to restore the original information. The integrity verification unit synchronously calls the national cryptographic SM3 hash algorithm each time it reads pipeline static information from an RFID tag or database. It calculates the hash value of the read information and compares it with the original hash value generated and securely stored when it was initially written. If the two are completely consistent, the data is determined to be intact and has not been tampered with. If they are inconsistent, an alarm for data integrity verification failure is immediately issued to the system and the subsequent operation process is stopped.

[0019] In one embodiment, the digital asset ledger module includes: The information input and verification unit is used to receive the static information of the pipeline and to perform standardization and compliance verification on the static information of the pipeline. The ledger dynamic maintenance unit is used to link the static information of pipelines with real-time monitoring data using a unique identifier as an index, and to build a digital asset ledger. The comprehensive query unit, based on the digital asset ledger, supports users to perform combined queries based on multiple conditions and generate query results through multi-dimensional statistical analysis, and provides a visual display and export of the query results; Furthermore, the information entry and verification unit imports pipeline static information in batches through a predefined application programming interface via a web form interface or mobile application provided by the system. This unit has a built-in standardized verification rule library containing data types, numerical ranges, and encoding rules, automatically performing format and logic checks on the entered pipeline static information and comparing it with an industry standard database to complete compliance verification. For items that fail verification, the system will automatically mark them and trigger a manual review process. The ledger dynamic maintenance unit establishes an asset master table in the system's backend core database, with the pipeline's unique identifier as the primary key. This master table is associated and bound to the attribute table storing pipeline static information and the real-time monitoring data stream table received from the IoT gateway through unique identifiers. Through database triggers and message queue mechanisms, when new real-time monitoring data is received... When a lifecycle status change command is received, the system automatically updates dynamic fields such as current status, cumulative loss, and latest coordinates in the asset master table and generates a timestamped asset change log to ensure the dynamism and consistency of the digital asset ledger. The comprehensive query unit provides a graphical query interface on the front end that includes drop-down filtering, range selection, and keyword search. After a user submits a query request, the backend service dynamically generates a structured query statement based on the query request and performs pre-aggregation and multi-dimensional modeling on massive ledger data at the data warehouse level to support complex statistical analysis. The query and statistical results are rendered into visualizations such as line charts, bar charts, or asset distribution maps through an integrated Gaussian chart library. It also provides a one-click export function to portable document or spreadsheet formats to meet diverse management, reporting, and auditing needs.

[0020] In one embodiment, the lifecycle management module includes: The lifecycle start-of-life recording unit creates a complete asset entry for the corresponding pipeline in the digital asset ledger based on the pipeline static information obtained after decrypting the unique identifier. The process recording unit is used to record the deployment location and operation parameters of the pipeline as real-time monitoring data based on the unique identifier identified during the pipeline's commissioning process, and to update the real-time monitoring data to the digital asset ledger. The status transition unit is used to update the status identifier of the pipeline in the digital asset ledger based on the inspection results after the pipeline recycling inspection. The scrapping process management unit is used to execute an online approval process for pipelines whose status is marked as pending scrapping, and after the approval is completed, update their status to scrapped in the digital asset ledger and attach an asset freeze mark. The pipeline lifecycle management module outputs full lifecycle data of the pipeline, which includes static pipeline information, real-time monitoring data, and status indicators. Furthermore, the lifecycle start-up recording unit calls the data encryption / decryption unit and uses the corresponding national cryptographic SM4 algorithm key to decrypt the ciphertext of the unique identifier read from the RFID tag, restoring the pipeline's static information. Then, it automatically creates a new record in the digital asset ledger's data table with this unique identifier as the primary key, and fills the corresponding fields with the decrypted production information and other pipeline static information as initial values, completing the initialization of the asset entry. The usage process recording unit scans the tags on the pipeline at the well site using an industrial-grade handheld terminal or fixed reader to identify their unique identifiers. The operator selects or inputs parameters such as the well number, well depth, current operating pressure, and geological strata on the terminal application. The terminal binds these parameters with a timestamp and the unique identifier and transmits them to the server via an encrypted network. The server receives this data and inserts it in real-time into the monitoring log table associated with the asset master table, while simultaneously updating the current location and latest operating status fields in the asset master table. The status transition unit records the pipeline's recovery to... After arriving at the base, inspectors scan the label and select preset inspection conclusion options on the inspection terminal, such as "intact and ready for use," "minor wear and requiring repair," or "severe corrosion." This conclusion serves as a new status identifier, automatically updating the status field of the pipeline record in the digital asset ledger through the business logic layer and triggering relevant notifications. When the status identifier is updated to "pending scrapping," the scrapping process management unit automatically generates a multi-level electronic approval process in the office automation system, linking department heads, financial asset administrators, and supervisors. Each approval node authenticates and makes decisions through digital signatures. After all approval nodes have passed, the system automatically executes the final update operation, changing the pipeline's status identifier in the ledger to "scrapped" and setting an asset freeze flag boolean value to true. This flag will prevent any subsequent outbound or downhole operation instructions for this pipeline. Finally, this module provides an integrated pipeline lifecycle data view through a database view or application programming interface, which integrates the pipeline static information table, real-time monitoring data flow table, and current status identifier table, ensuring data integrity and consistency.

[0021] In one embodiment, the intelligent analysis module includes: The digital profile building unit acquires pipeline lifecycle data and builds a dynamically updated multi-dimensional feature vector for each pipeline based on the pipeline lifecycle data, which serves as the digital profile of the pipeline. The pipeline intelligent agent modeling unit, based on digital profiling, creates a pipeline intelligent agent model with autonomous decision-making logic for each pipeline. The pipeline intelligent agent model includes the digital profiling of the pipeline and the environmental adaptability evaluation function. The virtual environment interaction unit is used to construct a virtual interaction environment that simulates the target downhole working conditions, and to put multiple candidate pipeline intelligent agent models into the virtual interaction environment for collaborative operation simulation. The group collaborative optimization unit is used to simulate and evaluate the interaction of different pipeline combinations in a virtual environment and generate the optimal pipeline or pipeline combination scheme. The optimal piping or piping combination scheme will be presented in the analysis report; Furthermore, the digital profile construction unit obtains structured pipeline lifecycle data from the system's data platform via a database connection interface. It then extracts static pipeline information, time-series real-time monitoring data, and status identification records as input sources. Feature extraction algorithms are used to clean the data and calculate key statistical features, such as historical average pressure, maximum corrosion rate, and fatigue cumulative coefficients under different operating conditions. Principal component analysis and other dimensionality reduction techniques are then used to fuse and normalize high-dimensional features, ultimately generating a digital profile for each pipeline represented by a numerical vector, which dynamically updates with new monitoring data. The pipeline intelligent agent modeling unit uses this digital profile as the core state input for the intelligent agent. First, it initializes an intelligent agent framework with an internal state machine based on features such as material strength and historical damage contained in the profile. Then, a pre-trained multi-layer neural network model is used to construct an environmental fitness evaluation function. The training data for this neural network comes from the mapping relationship between the pipeline's historical monitoring data and corresponding environmental parameters, as well as performance labels derived from status identification. The trained function enables the intelligent agent to output a quantified fitness score upon receiving environmental state input. Simultaneously, the unit configures a set of preset behaviors for each intelligent agent. Strategy rules, such as triggering a risk flag when the simulated pressure exceeds a certain proportion of the material's yield threshold, or prioritizing negotiation and load sharing with adjacent agents of similar wall thickness in load allocation; the virtual environment interaction unit, based on the target well's engineering design, digitally models key operating parameters such as well depth, formation pressure, temperature, and hydrogen sulfide concentration, and constructs a computable pressure and temperature field model based on simplified principles of computational fluid dynamics and solid mechanics as the physical core of the virtual interaction environment. Subsequently, the spatial relationships of multiple pipeline agent models to be evaluated are initialized as environment instances according to the pipe string combination scheme, enabling group collaboration. The optimization unit drives the virtual environment instance to perform iterative calculations at time steps. Within each time step, each pipeline agent model calculates the overall environmental fitness based on the current environmental pressure and temperature state through its environmental fitness evaluation function. The combination sequence and spatial arrangement of the agent models are continuously adjusted through the traversal optimization algorithm. Based on the average fitness, worst fitness, and risk consistency index of the entire group of agents after multiple rounds of simulation, the global performance of the combination scheme is comprehensively evaluated. Finally, the pipeline agent model combination scheme with the best global performance index is output as an analysis report, which includes a list of entity pipeline numbers and their arrangement order.

[0022] In one embodiment, the step of creating a pipeline intelligent agent model with autonomous decision-making logic for each pipeline based on digital profiling, wherein the pipeline intelligent agent model includes a digital profiling of the pipeline and an environmental adaptability evaluation function, is as follows: Based on the pipeline's entire lifecycle data, extract the pipeline's static information, real-time monitoring data, and status indicators; Based on feature extraction algorithms, the pipeline static information, real-time monitoring data and status indicators are normalized and feature encoded to generate multi-dimensional feature vectors as digital profiles. The digital profile is used as the initial model for the pipeline agent model, and behavioral policies are applied to the initial model to generate the pipeline agent model. Among them, the behavioral strategies include deformation under environmental pressure loads, stress response thresholds, and coordination rules with other pipeline agents in load distribution. The training sample set is formed by using real-time monitoring data and its corresponding environmental parameters as model input features and the performance degradation index derived from the state identifier sequence as training target label. The training sample set is input into the neural network model to generate the environmental fitness evaluation function of the pipeline agent model. Furthermore, complete pipeline lifecycle data for the target pipeline is extracted from the system's data warehouse via a database query interface. This includes pipeline static information, real-time monitoring data sorted by timestamp, and historical status identification records. Then, feature extraction algorithms are used to fuse these three types of data. Categorical data such as material type in the pipeline static information is thermally encoded. The mean, variance, extreme values, and trend characteristics of numerical sequences such as pressure and temperature in the real-time monitoring data are calculated. Status identification records are sequentially encoded. Finally, all features are normalized to eliminate dimensional differences. The final result is a unified high-dimensional feature vector representing the pipeline. A digital profile is generated; this digital profile is loaded as initialization parameters into a pre-defined intelligent agent framework. The intelligent agent framework defines the internal state and decision-making interface of the agent and applies pre-defined behavioral policies to it. The behavioral policies exist in the form of a rule base, such as defining a risk flag to be triggered when the simulated stress value exceeds a certain proportion of the material yield strength in the digital profile, and a collaborative rule to negotiate and share the excess load proportionally based on the wall thickness characteristics when communicating with neighboring intelligent agents in the virtual environment. An environmental fitness evaluation function is constructed through supervised learning. For supervised learning, multiple sets of real-time monitoring data and their corresponding well depths and formation pressures are extracted from historical data during training. Environmental parameters such as force are used as input features, and the state identifier sequence of the same period is quantified into a continuous remaining life percentage or failure risk coefficient as target labels through a predefined degradation model to form a training sample set. Finally, the sample set is input into a multi-layer feedforward neural network model for training. This network uses the backpropagation algorithm to optimize the weights. After training, the network is fixed as the environmental fitness evaluation function of the pipeline agent model. It can calculate a scalar score representing its fitness degree for new environmental state inputs. For example, when constructing its environmental fitness evaluation function for a specific steel grade oil pipe, the training process extracts data from the historical service records of the oil pipe. Sample data from multiple well runs are collected. The input features of each sample integrate real-time monitoring data of the well run process, such as average well depth, average operating pressure, and average temperature, as well as the specific downhole environmental parameters corresponding to these data. The status identification transformation sequence of the pipeline is also collected. For example, after the first recovery, it becomes slightly corroded, and it may evolve into moderate fatigue in subsequent stages. The status identification transformation sequence is quantified into continuous numerical indicators through a preset quantification model. That is, the new pipe status is mapped to 100% remaining life, and the first occurrence of slightly corroded status is mapped to 80%. Thus, the specific remaining life percentage value after the well run is obtained as the training target label for this sample.The input features of all historical samples are combined with their corresponding target labels to form a training sample set, which is then fed into a multilayer feedforward neural network model for training. After training, this network becomes the environmental fitness evaluation function for this specific pipeline. When it is necessary to evaluate the adaptability of this pipeline in a new well, such as a well with a depth of 4000 meters and a pressure of 70 MPa, the downhole environmental parameters of that well are input into the trained multilayer feedforward neural network model, which can then output a quantified fitness score.

[0023] In one embodiment, the step of constructing a virtual interactive environment simulating the target downhole working condition and placing multiple candidate pipeline intelligent agent models into the environment for collaborative operation simulation is as follows: Based on the design requirements of the target downhole conditions, key environmental parameters are extracted, including well depth, formation pressure, temperature gradient, and concentration of corrosive components in the medium. Key environmental parameters are normalized to generate the initial input conditions for the virtual interactive environment. Input the initial input conditions into the preset physical rule model to generate a virtual interactive environment; Multiple pipeline agent models are transmitted to a virtual interactive environment to generate virtual interactive environment instances. The virtual interactive environment instance is driven to run according to the preset operation sequence, and the behavior strategies and environmental adaptability evaluation functions of all pipeline intelligent agent models are activated synchronously to simulate collaborative operation. Furthermore, firstly, detailed operating parameters of the target well are extracted from the drilling engineering design documents or database of the oilfield, including well depth, formation pressure profile, geothermal gradient, and concentration data of corrosive components such as hydrogen sulfide and carbon dioxide in the drilling fluid. Simultaneously, based on real-time acquisition of dynamic environmental parameters from multi-source sensors pre-deployed downhole, the detailed operating parameters and dynamic environmental parameters are standardized using a maximum-minimum normalization method, scaling them to a common numerical range of [0,1] to generate initial input conditions in a unified format required for the virtual interactive environment. These initial input conditions are then imported into a pre-defined physical rule model constructed based on the finite element method and chemical kinetic equations. This model calculates the internal dynamics of the wellbore by solving simplified Navier-Stokes equations. The system simulates temperature field changes by analyzing pressure and flow velocity distribution using the heat conduction equation and calculates material loss at different locations using a corrosion rate model, dynamically generating a digital twin virtual interactive environment with spatial gradients and temporal evolution. Based on this virtual interactive environment, the system selects multiple candidate pipeline agent models from the pipeline agent model library according to preliminary screening rules. Through the application programming interface, it loads their internal states, including digital profiles and corresponding environmental fitness evaluation functions, onto the specified three-dimensional spatial coordinate nodes of the virtual interactive environment, completing the assembly of the virtual interactive environment instance. The preliminary screening rules first generate a set of computable screening rules based on the key constraints of the target environment. Subsequently, it sequentially traverses all constructed models in the pipeline agent model library. For each instance, a multi-level rule matching process is executed sequentially. The first-level rule, based on the latest status identifier recorded in the digital asset ledger for the pipeline corresponding to the model, filters only available pipeline models with status identifiers of idle or intact and ready for use, excluding pipeline agent models in states such as being in the well, awaiting maintenance, or scrapped. The second-level rule calls the digital profile encapsulated by the pipeline agent model and compares its inherent attribute characteristics with the target operating condition requirements. For example, it filters models whose steel grade characteristic value in the digital profile is greater than or equal to the target well design pressure requirement threshold, and whose material sulfur resistance grade characteristic matches the concentration of corrosive media in the target environment. The third-level rule matches the pipeline agent model's own historical evaluation records with the target environment, and the system checks... The model is queried based on its past simulation records. If the features represented by its digital profile show a high fitness evaluation value in similar environmental parameters, it is considered a candidate. Finally, the set of pipeline agent models that pass all screening rules is selected as candidate pipeline agent models. Ultimately, the unit drives the virtual interactive environment instance through a discrete-time step-by-step simulation engine, gradually applying simulated loads according to the temporal logic of actual operations. At each time step, the computational kernels of all pipeline agent models in the environment are activated synchronously, allowing them to call their own behavioral strategies to make decisions based on the current local environmental state and run their environmental fitness evaluation functions in real time to update their own state scores. This achieves dynamic interaction and collaborative operation simulation of multiple agents under simulated working conditions.

[0024] In one embodiment, the step of simulating and evaluating the interaction of different pipe combinations in a virtual environment to generate the optimal pipe or pipe combination scheme is as follows: In a virtual interactive environment, the pipeline intelligent agent model that drives the input performs multi-round collaborative operation simulations based on behavioral strategies and environmental fitness evaluation functions; Based on the real-time performance of multi-round collaborative operation simulation, an overall performance index is generated to evaluate different combinations. A target optimization algorithm is used to iteratively adjust the combination of pipeline agent models; When the overall performance index reaches its optimal level, the corresponding pipeline intelligent agent model is combined and the output is the optimal pipeline or pipeline combination scheme. Furthermore, this unit drives the virtual interactive environment instance through a discrete-time step-by-step simulation engine. In each simulation round, the engine gradually applies the dynamic loads calculated by the virtual interactive environment according to a preset operation sequence, and simultaneously activates all pipeline agent models in the environment. Each pipeline agent model makes decisions based on real-time acquired local environmental state data, invokes its own behavioral strategy, and runs its environmental fitness evaluation function to calculate the instantaneous fitness score. The system continuously collects the fitness scores and interaction decision records of each model at each time step as real-time performance data for evaluation. Based on the dataset collected from multiple complete simulation cycles, the unit calculates the overall performance index of the current pipeline agent model combination through a weighted aggregation function. This function comprehensively considers the average and variance of the fitness scores of all models to evaluate the overall efficiency and risk. The system considers risk balance and specifically incorporates the lowest fitness score of the key model as a bottleneck constraint index into the calculation. Then, the unit uses an ergonomic optimization algorithm to systematically iteratively try all possible combinations composed of candidate pipeline agent models. The combination is then placed in a virtual interactive environment for rapid simulation to calculate its new overall performance index. The system records and compares the overall performance index obtained in each round of attempts, and this process is repeated iteratively. Finally, when all predefined candidate combinations have been tried and evaluated, or when the overall performance index has reached the preset optimal threshold, the iteration process terminates. At this point, the unit selects the set of pipeline agent models with the best overall performance index from all the tried combinations, and outputs the list of unique identifiers of the entity pipelines corresponding to this model set and their spatial arrangement order in the virtual interactive environment as the final optimal pipeline or pipeline combination scheme.

[0025] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A pipeline lifecycle management system based on MA identifiers and national cryptographic algorithms, characterized in that, Data acquisition and security processing module: Collects static pipeline information for each pipeline and marks it with a unique identifier. Based on the national commercial cryptographic algorithm, it securely encrypts the unique identifier and its associated static pipeline information. Digital Asset Ledger Module: Connected to the data acquisition and security processing module, it uses unique identifiers as indexes to build a digital asset ledger for the pipeline; Lifecycle Management Module: Connects to the Digital Asset Ledger Module to track the status of pipelines at each stage from production to disposal, generate pipeline status tracking records, and transmit the status tracking records to the Digital Asset Ledger for updating the Digital Asset Ledger and generating full lifecycle data of the pipelines. Intelligent Analysis Module: Connects to the lifecycle management module, and based on the pipeline's full lifecycle data, predicts the pipeline's status and generates pipeline analysis reports.

2. The pipeline lifecycle management system based on MA identifier and national cryptographic algorithm according to claim 1, characterized in that, The data acquisition and security processing module includes: The identification encoding and writing unit is used to obtain the pipeline MA unified identification code and its static information from the associated system, and write the MA unified identification code and its static information as a unique identifier to the physical carrier attached to the pipeline. The data encryption and decryption unit is used to encrypt and decrypt unique identifiers by calling national commercial cryptographic algorithms during data transmission and storage. The integrity verification unit is used to verify the integrity of pipeline static information based on national commercial cryptographic algorithms during data reading.

3. The pipeline lifecycle management system based on MA identifier and national cryptographic algorithm according to claim 1, characterized in that, The digital asset ledger module includes: The information input and verification unit is used to receive the static information of the pipeline and to perform standardization and compliance verification on the static information of the pipeline. The ledger dynamic maintenance unit is used to link the static information of pipelines with real-time monitoring data using a unique identifier as an index, and to build a digital asset ledger. The comprehensive query unit, based on the digital asset ledger, allows users to perform combined queries based on multiple conditions and generate query results through multi-dimensional statistical analysis. The query results can also be visualized and exported.

4. The pipeline lifecycle management system based on MA identifier and national cryptographic algorithm according to claim 1, characterized in that, The lifecycle management module includes: The lifecycle start-of-life recording unit creates a complete asset entry for the corresponding pipeline in the digital asset ledger based on the pipeline static information obtained after decrypting the unique identifier. The process recording unit is used to record the deployment location and operation parameters of the pipeline as real-time monitoring data based on the unique identifier identified during the pipeline's commissioning process, and to update the real-time monitoring data to the digital asset ledger. The status transition unit is used to update the status identifier of the pipeline in the digital asset ledger based on the inspection results after the pipeline recycling inspection. The scrapping process management unit is used to execute an online approval process for pipelines whose status is marked as pending scrapping, and after the approval is completed, update their status to scrapped in the digital asset ledger and attach an asset freeze mark. The pipeline lifecycle management module outputs full lifecycle data of the pipeline, which includes static information, real-time monitoring data and status indicators.

5. The pipeline lifecycle management system based on MA identifier and national cryptographic algorithm according to claim 1, characterized in that, The intelligent analysis module includes: The digital profile building unit acquires pipeline lifecycle data and builds a dynamically updated multi-dimensional feature vector for each pipeline based on the pipeline lifecycle data, which serves as the digital profile of the pipeline. The pipeline intelligent agent modeling unit, based on digital profiling, creates a pipeline intelligent agent model with autonomous decision-making logic for each pipeline. The pipeline intelligent agent model includes the digital profiling of the pipeline and the environmental adaptability evaluation function. The virtual environment interaction unit is used to construct a virtual interaction environment that simulates the target downhole working conditions, and to put multiple candidate pipeline intelligent agent models into the virtual interaction environment for collaborative operation simulation. The group collaborative optimization unit is used to simulate and evaluate the interaction of different pipeline combinations in a virtual environment and generate the optimal pipeline or pipeline combination scheme. The optimal piping or piping combination scheme will be presented in the analysis report.

6. The pipeline lifecycle management system based on MA identifier and national cryptographic algorithm according to claim 5, characterized in that, The steps for creating a pipeline intelligent agent model with autonomous decision-making logic for each pipeline based on digital profiling, whereby the pipeline intelligent agent model includes the pipeline's digital profiling and an environmental adaptability evaluation function, are as follows: Based on the pipeline's entire lifecycle data, extract the pipeline's static information, real-time monitoring data, and status indicators; Based on feature extraction algorithms, the pipeline static information, real-time monitoring data and status indicators are normalized and feature encoded to generate multi-dimensional feature vectors as digital profiles. The digital profile is used as the initial model for the pipeline agent model, and behavioral policies are applied to the initial model to generate the pipeline agent model. Among them, the behavioral strategies include deformation under environmental pressure loads, stress response thresholds, and coordination rules with other pipeline agents in load distribution. The training sample set is formed by using real-time monitoring data and its corresponding environmental parameters as model input features and the performance degradation index derived from the state identifier sequence as training target label. The training sample set is input into the neural network model to generate the environmental fitness evaluation function of the pipeline agent model.

7. The pipeline lifecycle management system based on MA identifier and national cryptographic algorithm according to claim 5, characterized in that, The steps for constructing a virtual interactive environment that simulates the target downhole working conditions and placing multiple candidate pipeline intelligent agent models into this environment for collaborative operation simulation are as follows: Based on the design requirements of the target downhole conditions, key environmental parameters are extracted, including well depth, formation pressure, temperature gradient, and concentration of corrosive components in the medium. Key environmental parameters are normalized to generate the initial input conditions for the virtual interactive environment. Input the initial input conditions into the preset physical rule model to generate a virtual interactive environment; Multiple pipeline agent models are transmitted to a virtual interactive environment to generate virtual interactive environment instances. The virtual interactive environment instance is driven to run according to the preset operation sequence, and the behavior strategies and environmental adaptability evaluation functions of all pipeline intelligent agent models are activated synchronously to simulate collaborative operations.

8. The pipeline lifecycle management system based on MA identifier and national cryptographic algorithm according to claim 5, characterized in that, The steps for simulating and evaluating the interaction of different pipe combinations in a virtual environment to generate the optimal pipe or pipe combination scheme are as follows: In a virtual interactive environment, the pipeline intelligent agent model that drives the input performs multi-round collaborative operation simulations based on behavioral strategies and environmental fitness evaluation functions; Based on the real-time performance of multi-round collaborative operation simulation, an overall performance index is generated to evaluate different combinations. A target optimization algorithm is used to iteratively adjust the combination of pipeline agent models; When the overall performance index reaches its optimal level, the corresponding pipeline intelligent agent model is combined, and the output is the optimal pipeline or pipeline combination scheme.

Images