Construction procedure driven oil and gas field digital twin dynamic growth and verification method

Abstract

This invention discloses a method for dynamic growth and verification of digital twins for oil and gas fields driven by construction procedures, belonging to the field of digital oil and gas field engineering. The method first constructs a digital twin foundation based on oil and gas field engineering design data, then defines the construction procedure sequence and configures incremental data packages and triggering conditions for deliverables. Dynamic growth of the twin is achieved driven by procedure completion events, while simultaneously updating construction change data back to the design database. Subsequently, closed-loop verification is performed based on calculations and measurements. When deviations exceed limits, iterative updates of model parameters are triggered. Finally, the procedure configuration, dynamic growth, and verification update steps are executed cyclically as construction procedures progress, achieving consistent maintenance of the digital twin throughout its entire lifecycle. This invention solves the problems of static digital twin models, difficulty in backflowing change data, and fragmented data organization in traditional oil and gas field models, improving the consistency and reliability of the digital twin and the physical entity throughout their entire lifecycle, and is applicable to the digital management of the entire lifecycle of oil and gas field engineering.

Description

Technical Field

[0001] This invention relates to the field of digital technology for oil and gas field engineering, and in particular to a method for dynamic growth and verification of digital twins of oil and gas fields driven by construction procedures. Background Technology

[0002] With the continuous expansion of oil and gas field engineering construction and the increasing demand for digital and intelligent transformation in the oil and gas industry, digital twin technology, as an important means to achieve full life-cycle management of oil and gas fields, has received widespread attention in recent years in areas such as oil and gas field surface engineering construction, pipeline operation monitoring, production optimization, and safety management. Digital twins typically refer to the construction of a digital model corresponding to a physical entity in virtual space, and the mapping, simulation, prediction, and decision support of the entity's state driven by multi-source data.

[0003] In existing technologies, the construction of digital twins for oil and gas field projects typically relies on Building Information Modeling (BIM) and Geographic Information System (GIS) technologies. During the design phase, BIM is used to create detailed 3D models of the site's internal equipment, pipelines, and building structures, while GIS is combined to construct the external geospatial environment, thus achieving integrated indoor and outdoor spatial representation and visualization. Related research mainly focuses on the fusion and conversion of BIM and GIS data models, spatial registration, and scene integration. However, these solutions often emphasize the static construction of the digital foundation, with the model deliverables typically completed and delivered only during the design phase, lacking a mechanism for dynamic evolution as the construction progresses. When changes occur during construction, such as pipeline rerouting, equipment replacement, or parameter adjustments, the virtual model struggles to be updated in a timely manner, causing the digital twin to gradually deviate from the physical entity, making it difficult to guarantee consistency throughout the entire lifecycle.

[0004] To further reflect the time dimension information of the construction phase, some existing technologies link 3D BIM models with construction schedules to form so-called 4D BIM models, used for progress simulation, phase display, and plan comparison analysis during construction. Simultaneously, some studies combine point cloud scanning and visual recognition to automatically measure construction progress at the site and compare it with the planned model, thus achieving a certain degree of construction progress monitoring. However, these solutions typically still focus on the planned schedule, primarily showcasing the construction process through visualization or phased comparisons, and struggle to reflect the real-time status information of multiple elements such as "people, machinery, and materials" at the construction site. Furthermore, the large amount of change data generated during construction often lacks an effective mechanism for structured writing back to the design model and twin database, resulting in incomplete and inconsistent model information after completion and delivery, affecting data inheritance and asset management in subsequent operation and maintenance phases.

[0005] During the production and operation phase of oil and gas fields, existing technologies generally collect real-time operating parameters such as pressure, temperature, and flow rate through IoT sensors, and map the monitoring data onto virtual models for status display, early warning analysis, or simulation prediction. Some studies have proposed combining the Industrial Internet of Things (IIoT) with digital twins to construct a closed-loop framework integrating "simulation calculation - field measurement" to improve the accuracy of model prediction and analysis. In addition, some patents disclose construction progress management methods and systems based on digital twins, achieving progress management by associating construction plans with models. However, the above solutions are mostly focused on monitoring or updating and displaying construction progress during the operation and maintenance phase. Their data systems are often independent of the engineering construction phase, lacking a unified data organization structure that runs through design, construction, and operation and maintenance; they also lack a mechanism for releasing construction change data back to the design phase, as well as an asset-based data management approach oriented towards physical objects. Due to the long and variable construction cycle of oil and gas field projects, and the evolution of model parameters over time, existing technologies generally lack the ability to continuously verify and dynamically update parameters driven by multi-source real-time data, resulting in insufficient credibility and consistency of digital twins.

[0006] In summary, existing digital twin technologies for oil and gas fields still suffer from problems in practical engineering applications, such as static model construction, difficulty in backtracking construction changes, fragmented data organization, lack of closed-loop verification, and inability to support continuous evolution throughout the entire lifecycle. Therefore, there is an urgent need for a technical solution that can drive the dynamic growth of the digital twin through construction procedures and achieve consistency between the digital twin and the physical oil and gas field engineering throughout its entire lifecycle through construction data backtracking and closed-loop verification using multi-source real-time data. Summary of the Invention

[0007] This invention addresses the shortcomings of existing digital twin technologies for oil and gas fields by providing a process-driven dynamic growth and verification method for digital twins of oil and gas fields. This method achieves high consistency and high reliability between the digital twin and the physical oil and gas field engineering entity throughout the entire lifecycle, solving problems such as static model delivery, difficulty in reverting changed data, fragmented data organization, lack of closed-loop verification of simulation models, and difficulty in supporting the evolution and reuse of models throughout the entire lifecycle.

[0008] To solve the above-mentioned technical problems, the present invention adopts the following solution: A method for dynamic growth and verification of digital twins in oil and gas fields driven by construction procedures includes the following steps: S1. Input the basic design data of oil and gas field engineering into the digital twin platform, complete the initialization of the digital twin, the splitting and encoding of entity objects based on the entity object decomposition structure, the construction of multi-dimensional sub-models, and output the digital twin base of the oil and gas field. S2. Input the oil and gas field digital twin base and construction procedure planning data into the digital twin platform. Define the construction procedure sequence and configure the incremental data package of deliverables and procedure triggering conditions within the digital twin platform. Output the construction procedure sequence and incremental data package configuration scheme with procedure triggering conditions. The procedure triggering conditions are used to determine whether the construction procedure is completed and trigger the dynamic growth of the digital twin. The procedure triggering conditions include receiving the electronic file generated by the procedure acceptance record, obtaining the online confirmation instruction of the construction visa, receiving the input signal containing material scanning information and installation completion status, or the collected test data meeting the specification requirements. S3. Input the construction process sequence with process triggering conditions, the incremental data package configuration scheme of deliverables, and the process completion data at the construction site into the digital twin platform. Perform incremental mounting within the digital twin platform to achieve dynamic growth of the twin. At the same time, record construction changes and complete the backflow update to the main database of the design stage. Output the updated digital twin and the synchronously updated main database of the design stage. S4. Input the updated digital twin and IoT real-time monitoring data into the digital twin platform, collect the simulation model output data in the digital twin platform and perform calculation and measurement closed-loop verification, output the calculation and measurement deviation results, and output the model parameter iterative update trigger command when the deviation exceeds the preset threshold. S5. Input the model parameter iterative update trigger command and digital twin model parameters into the digital twin platform. The digital twin model parameters include wellbore friction factor, geostress coefficient, flow resistance coefficient, heat transfer coefficient, and equipment performance curve parameters. Perform model parameter iterative updates and automatic verification of the six properties of twin data within the digital twin platform. The automatic verification of the six properties of twin data is achieved by configuring the verification requirements as executable verification rules and associating them with the process trigger conditions. When the process is completed or the digital twin data is updated, the corresponding verification rules are automatically invoked to complete the verification. As the construction process progresses, new construction process data and on-site process completion data are cyclically input into the digital twin platform. Steps S2 to S4 are repeated to output a dynamically updated oil and gas field digital twin with a full life cycle.

[0009] Furthermore, in step S1, the initialization of the digital twin involves constructing the digital twin into a six-tuple that includes a set of oil and gas field engineering entity objects, a set of multi-dimensional sub-models corresponding to the entity objects, a construction procedure sequence, a design phase master database, a construction phase release database, and a set of twin consistency verification functions.

[0010] Specifically, the twin consistency check function is used to compare the consistency of design phase data and construction phase data in terms of geometry, parameters and correlation, and to calculate the change deviation. When the change deviation is zero, the design data and construction data are determined to be consistent. When the change deviation is non-zero, the construction change is determined to exist, and the construction change data is triggered to be updated.

[0011] Furthermore, in step S1, the entity object splitting and encoding based on the entity object decomposition structure involves splitting the entity into the smallest entity units of pipe sections, valves, structures, and oil and gas field process equipment, and generating a unique identity code for each smallest entity unit, which is processed by a hash mapping function based on the object category, spatial location code, equipment tag number, or pipeline number; the multi-dimensional sub-model set includes BIM three-dimensional geometric model, GIS geospatial model, flow thermodynamics simulation model, and entity object attribute state relationship data model.

[0012] Furthermore, in step S2, the incremental data package of deliverables is configured in the digital twin platform as a structured data package bound to the unique identity code of the entity object, including the process deliverables, the process completion timestamp, and the version number; the process deliverables include the model, parameters, or document deliverables generated by the process.

[0013] Furthermore, in step S3, incremental mounting is performed within the digital twin platform. This involves mounting the incremental data package of the results to the digital twin base of the oil and gas field using an incremental mounting operator. The incremental mounting operator can be implemented in one or more of the following ways: attribute appending, parameter overwrite update, establishing entity object relationship edges, and generating new model version references. Attribute appending is one implementation of the incremental mounting operator. It refers to writing the attribute information added during the construction phase into the entity object attribute set as a new field while keeping the unique identity code of the entity object unchanged, thereby achieving incremental updates of object attributes.

[0014] Furthermore, in step S3, recording construction changes and completing the backflow update to the design phase master database involves first recording the incremental changes generated during the construction phase through the construction phase release database. Then, the consistency check function of the twin is used to compare the consistency between the design data and the construction data in terms of geometry, parameters, and relationships to calculate the change deviation. When the change deviation is non-zero, the construction change data is structured and written back to the design phase master database.

[0015] Furthermore, when version conflicts occur when construction change data is written back to the main database during the design phase, the following strategies are adopted: retaining both versions and marking their priorities, overwriting the design version with the actual construction version, or merging the versions after triggering manual confirmation.

[0016] Furthermore, in step S4, the real-time IoT monitoring data includes engineering operating parameters such as pressure, temperature, and flow rate; performing closed-loop verification within the digital twin platform involves comparing the real-time IoT monitoring data with the simulation model output data to calculate the measurement deviation E. t Set the error threshold as θ, if E t If E ≤ θ, the verification passes; if E tIf the value is greater than θ, it triggers an iterative update of the model parameters; the measurement method for the calculated deviation is any one of Euclidean distance, Mahalanobis distance, or relative error function.

[0017] Furthermore, in step S5, the iterative update of model parameters performed within the digital twin platform is to iteratively update the model parameters by combining the learning rate and the gradient of the loss function. The iteration termination condition is any one of the following: the measurement deviation is less than or equal to a preset threshold, the number of iterations reaches a preset upper limit, or the error change meets the convergence condition. The parameter update algorithm is any one of Kalman filtering, Bayesian update, or least squares estimation.

[0018] Furthermore, in step S5, the automatic verification of the six properties of the twin data executed within the digital twin platform is achieved by configuring the verification requirements of standardization, completeness, accuracy, consistency, timeliness, and accessibility as executable verification rules and associating them with the process triggering conditions. When the process is completed or the digital twin data is updated, the digital twin platform automatically calls the corresponding verification rules to automatically verify the encoding format, entity objects and their relationships, design data and measured data, database synchronization status, update time and version, data retrieval and access permission status of the digital twin data. If the verification fails, data correction, data synchronization, model parameter updates, or relationship reconstruction are automatically executed according to the type of anomaly.

[0019] The beneficial effects of this invention are as follows: 1. This invention completes the standardized decomposition and unique coding of oil and gas field engineering entities based on the entity object decomposition structure, and constructs a unified data organization system that runs through design, construction and operation and maintenance. It realizes the unique binding of engineering data and entity objects throughout the entire life cycle, solves the problems of fragmented data organization and lack of asset management in existing technologies, and greatly improves data retrieval efficiency and cross-stage reuse capability.

[0020] 2. This invention defines the construction process sequence and configures incremental data packages of deliverables. Driven by the completion events of construction process on site, it uses incremental mounting operators to realize the dynamic growth of the digital twin with each construction process, replacing the traditional one-time static modeling method in the design stage. It reflects the construction progress and the latest status of the physical object in real time, effectively avoiding the problems of deviation and distortion between the twin and the physical entity caused by long construction cycles and many changes.

[0021] 3. This invention establishes a database for the construction phase to systematically record incremental construction changes. It determines the consistency between design and construction data by measuring the amount of change deviation, realizes the structured backflow update of construction change data to the main database in the design phase, and configures a multi-strategy version conflict handling mechanism. This solves the problems of difficulty in writing back construction changes and incomplete and inconsistent model information after completion, ensuring a high degree of data consistency between the design model and the completed entity.

[0022] 4. This invention constructs a multi-source real-time data-driven closed-loop verification mechanism that compares real-time IoT monitoring data with simulation model output data to calculate deviations. When the error exceeds the limit, it automatically triggers iterative updates of model parameters. Combining multiple iteration termination conditions and replaceable parameter update algorithms, it achieves adaptive correction of the digital twin model, solving the problems of model drift and lack of experimental verification of simulation results in existing technologies. It continuously maintains the high fidelity of the digital twin and improves the accuracy of predictive analysis and decision support.

[0023] 5. This invention performs six-dimensional automated verification of digital twin data during parameter iteration and update, ensuring data quality from six dimensions including standardization and completeness. At the same time, through the cyclical execution of process-driven growth, change recirculation, and verification updates during the construction process, the digital twin is continuously evolved during the construction phase, continuously updated consistently during the change phase, and continuously self-verified and corrected during the operation phase. This effectively supports the consistency maintenance and reliable delivery of oil and gas field engineering throughout its entire lifecycle, and significantly improves the consistency and reliability of the digital twin and the physical oil and gas field engineering throughout their entire lifecycle. Attached Figure Description

[0024] Figure 1 This is the overall flowchart of the method for dynamic growth and verification of digital twins in oil and gas fields driven by construction procedures according to the present invention.

[0025] Figure 2 This is a diagram of the six-element digital twin of this invention.

[0026] Figure 3 This is a diagram of the incremental data package of the results of this invention. Detailed Implementation

[0027] 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, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. 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.

[0028] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0029] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0030] Furthermore, for clarity and brevity, descriptions of well-known structures, functions, and configurations may have been omitted. Those skilled in the art will recognize that various changes and modifications can be made to the examples described herein without departing from the spirit and scope of this disclosure.

[0031] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0032] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0033] like Figure 1 As shown, Figure 1 This is a flowchart illustrating the overall process of the construction process-driven dynamic growth and verification method for oil and gas field digital twins according to the present invention. The construction process-driven dynamic growth and verification method for oil and gas field digital twins includes the following steps: S1. Input the basic design data of oil and gas field engineering into the digital twin platform, complete the initialization of the digital twin, the splitting and encoding of entity objects based on the entity object decomposition structure, the construction of multi-dimensional sub-models, and output the digital twin base of the oil and gas field. S2. Input the oil and gas field digital twin base and construction procedure planning data into the digital twin platform. Define the construction procedure sequence and configure the incremental data package of deliverables and procedure triggering conditions within the digital twin platform. Output the construction procedure sequence and incremental data package configuration scheme with procedure triggering conditions. The procedure triggering conditions are used to determine whether the construction procedure is completed and trigger the dynamic growth of the digital twin. The procedure triggering conditions include receiving the electronic file generated by the procedure acceptance record, obtaining the online confirmation instruction of the construction visa, receiving the input signal containing material scanning information and installation completion status, or the collected test data meeting the specification requirements. S3. Input the construction process sequence with process triggering conditions, the incremental data package configuration scheme of deliverables, and the process completion data at the construction site into the digital twin platform. Perform incremental mounting within the digital twin platform to achieve dynamic growth of the twin. At the same time, record construction changes and complete the backflow update to the main database of the design stage. Output the updated digital twin and the synchronously updated main database of the design stage. S4. Input the updated digital twin and IoT real-time monitoring data into the digital twin platform, collect the simulation model output data in the digital twin platform and perform calculation and measurement closed-loop verification, output the calculation and measurement deviation results, and output the model parameter iterative update trigger command when the deviation exceeds the preset threshold. S5. Input the model parameter iterative update trigger command and digital twin model parameters into the digital twin platform. The digital twin model parameters include wellbore friction factor, geostress coefficient, flow resistance coefficient, heat transfer coefficient, and equipment performance curve parameters. Perform model parameter iterative updates and automatic verification of the six properties of twin data within the digital twin platform. The automatic verification of the six properties of twin data is achieved by configuring the verification requirements as executable verification rules and associating them with the process trigger conditions. When the process is completed or the digital twin data is updated, the corresponding verification rules are automatically invoked to complete the verification. As the construction process progresses, new construction process data and on-site process completion data are cyclically input into the digital twin platform. Steps S2 to S4 are repeated to output a dynamically updated oil and gas field digital twin with a full life cycle.

[0034] like Figure 2 As shown, Figure 2 This is a six-tuple diagram of the digital twin of the present invention. In step S1, the digital twin is initialized by constructing the digital twin into a six-tuple containing a set of oil and gas field engineering entity objects, a set of multi-dimensional sub-models corresponding to the entity objects, a construction procedure sequence, a design phase master database, a construction phase release database, and a set of twin consistency verification functions.

[0035] Specifically, the twin consistency check function is used to compare the consistency of design phase data and construction phase data in terms of geometry, parameters and correlation, and to calculate the change deviation. When the change deviation is zero, the design data and construction data are determined to be consistent. When the change deviation is non-zero, the construction change is determined to exist, and the construction change data is triggered to be updated.

[0036] Furthermore, in step S1, the entity object splitting and encoding based on the entity object decomposition structure involves splitting the entity into the smallest entity units of pipe sections, valves, structures, and oil and gas field process equipment, and generating a unique identity code for each smallest entity unit, which is processed by a hash mapping function based on the object category, spatial location code, equipment tag number, or pipeline number; the multi-dimensional sub-model set includes BIM three-dimensional geometric model, GIS geospatial model, flow thermodynamics simulation model, and entity object attribute state relationship data model.

[0037] like Figure 3 As shown, Figure 3 This is a diagram of the incremental data package of the deliverables of the present invention. In step S2, the incremental data package of deliverables is configured in the digital twin platform as a structured data package bound to the unique identity code of the entity object, including the process deliverables, the process completion timestamp, and the version number; the process deliverables include the model, parameters, or document deliverables generated by the process.

[0038] Furthermore, in step S3, incremental mounting is performed within the digital twin platform. This involves mounting the incremental data package of the results to the digital twin base of the oil and gas field using an incremental mounting operator. The incremental mounting operator can be implemented in one or more of the following ways: attribute appending, parameter overwriting and updating, establishing entity object relationship edges, and generating new model version references.

[0039] Furthermore, in step S3, recording construction changes and completing the backflow update to the design phase master database involves first recording the incremental changes generated during the construction phase through the construction phase release database. Then, the consistency check function of the twin is used to compare the consistency between the design data and the construction data in terms of geometry, parameters, and relationships to calculate the change deviation. When the change deviation is non-zero, the construction change data is structured and written back to the design phase master database.

[0040] Furthermore, when version conflicts occur when construction change data is written back to the main database during the design phase, the following strategies are adopted: retaining both versions and marking their priorities, overwriting the design version with the actual construction version, or merging the versions after triggering manual confirmation.

[0041] Furthermore, in step S4, the real-time IoT monitoring data includes engineering operating parameters such as pressure, temperature, and flow rate; performing closed-loop verification within the digital twin platform involves comparing the real-time IoT monitoring data with the simulation model output data to calculate the measurement deviation E. t Set the error threshold as θ, if E t If E ≤ θ, the verification passes; if E t If the value is greater than θ, it triggers an iterative update of the model parameters; the measurement method for the calculated deviation is any one of Euclidean distance, Mahalanobis distance, or relative error function.

[0042] Furthermore, in step S5, the iterative update of model parameters performed within the digital twin platform is to iteratively update the model parameters by combining the learning rate and the gradient of the loss function. The iteration termination condition is any one of the following: the measurement deviation is less than or equal to a preset threshold, the number of iterations reaches a preset upper limit, or the error change meets the convergence condition. The parameter update algorithm is any one of Kalman filtering, Bayesian update, or least squares estimation.

[0043] Furthermore, in step S5, the automatic verification of the six properties of the twin data executed within the digital twin platform is achieved by configuring the verification requirements of standardization, completeness, accuracy, consistency, timeliness, and accessibility as executable verification rules and associating them with the process triggering conditions. When the process is completed or the digital twin data is updated, the digital twin platform automatically calls the corresponding verification rules to automatically verify the encoding format, entity objects and their relationships, design data and measured data, database synchronization status, update time and version, data retrieval and access permission status of the digital twin data. If the verification fails, data correction, data synchronization, model parameter updates, or relationship reconstruction are automatically executed according to the type of anomaly.

[0044] Specifically, the digital twin platform calls a pre-set data quality rule library to configure the verification requirements of standardization, completeness, accuracy, consistency, timeliness, and accessibility as executable verification rules and associate them with the triggering conditions of the construction process. When the construction process is completed or the twin data is updated, the digital twin platform automatically calls the corresponding verification rules to perform automatic verification of the six properties.

[0045] Among them, the standardization verification automatically checks whether the data conforms to the preset specifications through data encoding rules, naming rules, unit and storage format rules; the integrity verification checks whether there is missing data or unattached objects by traversing entity objects, topological relationships and related documents; the accuracy verification identifies abnormal data by comparing design data, measured data and geometric model data and combining preset thresholds; the consistency verification compares the consistency of corresponding data items between the main database in the design phase, the release database in the construction phase and the operation and maintenance system; the timeliness verification determines whether the data meets the real-time update requirements by checking the data update time, event response delay and version information; and the accessibility verification determines whether the data can be accessed normally by detecting data retrieval, traceability links and permission access status.

[0046] When any verification item fails, one or more exception handling procedures will be automatically executed based on the exception type, including data correction, data synchronization, model parameter update, or relationship reconstruction.

[0047] The present invention will now be described in further detail with reference to specific embodiments.

[0048] This invention provides a method for dynamic growth and verification of digital twins in oil and gas fields driven by construction procedures. This method uses the Entity Object Structure (EBS) as the data organization basis and construction procedure events as the driving force to achieve dynamic growth of the digital twin synchronously with the physical project. Through a change feedback mechanism in the published database and a closed-loop verification process of "calculation-measurement" using multi-source real-time data, the method achieves continuous updating and reliable consistency of the digital twin.

[0049] The method of this invention is executed sequentially in a digital twin platform according to the following steps: Step S1: Digital Twin Initialization and Unified Object Modeling First, the basic representation of the oil and gas field digital twin is constructed, defining the twin as a six-tuple: X DT = <E, M, P, DB design DB release , F validate > Where: X DT This is the mathematical representation of an oil and gas field digital twin, used to describe its data organization structure; E represents the set of oil and gas field engineering entity objects; M represents the set of multi-dimensional sub-models corresponding to the entity objects; P represents the construction sequence; DB design This refers to the main database during the design phase; DB release Indicates the database is published during the construction phase; F validate This represents the set of twin consistency verification functions.

[0050] Step S2: Entity object splitting and unique identification coding based on EBS The entities of oil and gas field engineering are decomposed into objects, and an entity object decomposition structure (EBS) is constructed: E={e1,e2,…,e n} in It refers to the smallest physical unit, such as a pipe section, valve, equipment, or structure. Generate a unique identification code for each entity object: Where: type is the object category; location is the spatial location code; tag is the device tag number or pipeline number; It is a hash mapping function; through the EBS encoding system, it achieves a unique binding between engineering data and entity objects throughout the entire lifecycle.

[0051] Step S3: Multidimensional Sub-model Decomposition and Digital Foundation Construction For each entity object e i Establish its multidimensional sub-model set M(e) i ): in: For BIM 3D geometric model; M gis For GIS geospatial models; M sim For simulation model; M data This is a data model of the attribute state relationship of entity objects; thus forming an integrated indoor and outdoor digital twin base; M(e i ) is the entity object e i The corresponding set of multidimensional sub-models.

[0052] Step S4: Defining the construction sequence and constructing dynamic growth events The construction process is abstracted into a sequence of construction procedures: P={p1,p2,…,p k} Where P represents the mathematical representation of the construction sequence, p k This represents the k-th construction step; Define an incremental data package for deliverables for each process: ΔD(p k )= <id(e i ),Artifact,Time,Version> Where: ΔD(p) k This represents the incremental data package of the deliverables for the k-th construction process. Artifact refers to the model, parameters, or document deliverables generated by this process; Time is the process completion timestamp; and Version is the version number. Process triggering conditions are used to determine whether a construction process is completed and to trigger the dynamic growth of the digital twin. Process triggering conditions include receiving an electronic file generated from the process acceptance record, obtaining an online confirmation instruction for the construction visa, receiving an input signal containing material barcode information and installation completion status, or the collected test data meeting the specification requirements.

[0053] Step S5: Dynamic growth of twins based on incremental mounting operators. When the process status meets the completion conditions: Status() represents the process status retrieval function, used to obtain the current status of a construction process or entity object. k This represents the k-th construction procedure, and Finished indicates the completion status of the procedure, meaning that the corresponding construction procedure has met the procedure triggering conditions.

[0054] The system performs dynamic twin growth operations: Wherein: S t S represents the operating state of the digital twin at time t. t+1 This indicates the operational state of the digital twin at time t+1 after it has been mounted. This represents the incremental mounting operator, ΔD(p k ) represents the incremental data package of the deliverables of the k-th construction process; incremental mounting operator Equivalent implementation methods include: attribute appending; parameter overriding and updating; establishing entity relationship edges; generating new model version references; thereby enabling the digital twin to grow dynamically step by step with the construction process, rather than being delivered statically and once.

[0055] Step S6: Publish database records and construction change data feedback. Establish a construction phase publication database to record incremental changes generated during the construction phase; Define the change deviation amount : Among them, F validate This represents a set of twin consistency verification functions used to compare the consistency between design data and construction data in terms of geometry, parameters, and relationships; DB design The main database (DB) during the design phase release This indicates that the database will be published during the construction phase. If δ=0, it means there are no changes and no reflow is triggered; if δ≠0, then a reflow update is performed. Wherein, ∪ represents merging the incremental data packages of deliverables corresponding to construction procedures into the main database DB of the design phase in a structured manner. design .

[0056] If version conflicts occur during the backflow process, the following equivalent strategies can be adopted: retain both versions and mark their priorities; overwrite with the version measured during construction; merge after triggering manual confirmation; and realize the structured back-writing of construction data to the design stage by publishing the database.

[0057] Step S7: Multi-source real-time data-driven computation-measurement closed-loop verification of IoT real-time monitoring data stream: Where Y(t) represents the set of real-time IoT monitoring data collected at time t, characterizing the real-time operating status of the oil and gas field entity, including multi-dimensional monitoring indicators such as pressure, temperature, and flow. The digital twin platform calls the set of simulation models M in the digital twin constructed in step S3. sim M sim It includes one or more of the following: flow simulation model, thermal simulation model, and mechanical simulation model. It takes the updated digital twin model parameters and the boundary conditions (such as inlet pressure, ambient temperature, etc.) corresponding to the IoT real-time monitoring data as input, and performs online solution through the built-in simulation calculation engine (such as well wall stability mechanical analysis calculation engine or process calculation model) to obtain the simulation model output Sim(t) of the corresponding entity object, corresponding monitoring parameters and corresponding sampling time. It is then used to perform calculation and measurement closed-loop comparison and verification with the IoT real-time monitoring data Y(t).

[0058] Calculate the measurement deviation: E t Measurement deviation refers to the weighted Euclidean distance or relative deviation between the measured values ​​of a physical entity and the simulated values ​​of a virtual model, used to characterize the degree of difference between the operating state of the digital twin model and the physical object; w i Sim represents the weight coefficient of the i-th monitoring indicator; m represents the number of monitoring indicators involved in the calculation; i (t) represents the simulation model output data corresponding to the i-th monitoring indicator at time t; Y i (t) represents the real-time IoT monitoring value corresponding to the i-th monitoring indicator at time t; The threshold for determining the measurement deviation is θ, and the branching conditions are as follows: If E t ≤θ, verification passed; If E t >θ triggers iterative parameter updates; This embodiment uses weighted absolute error to calculate the measurement deviation as a way to implement the relative error function; in other embodiments, the error measurement method can be equivalently replaced by Euclidean distance, Mahalanobis distance or relative error function.

[0059] Step S8: Dynamic parameter iterative update and automatic verification of six properties When the calculated deviation exceeds a preset threshold, the simulation model parameters in the digital twin are iteratively updated: in: and E represents the simulation model parameters before and after the update. t Indicates the measurement deviation. This indicates the configuration parameters corresponding to the parameter update algorithm. This represents the parameter update operator. In different implementations, the parameter update operator can be any of gradient descent, Kalman filtering, Bayesian update, or least squares estimation. Specifically, gradient descent updates the simulation model parameters based on the gradient of the loss function; Kalman filtering corrects the simulation model parameters based on the state estimate; Bayesian update corrects the simulation model parameters based on the posterior probability distribution; and least squares estimation updates the parameters by minimizing the residual between the simulation model output and the real-time IoT monitoring data.

[0060] The iteration termination condition includes: the calculated deviation E t ≤θ; The number of iterations reaches the upper limit N. max Or the error change satisfies any one of the convergence conditions.

[0061] After the model parameters are updated, the digital twin platform calls the preset data quality rule library to automatically verify the standardization, completeness, accuracy, consistency, timeliness, and accessibility of the twin data. When any verification item fails, one or more exception handling processes are executed according to the exception type, including data correction, data synchronization, model parameter update, or relationship reconstruction.

[0062] Step S9: Execute repeatedly and maintain consistency throughout the entire lifecycle. The above steps are repeated during the construction process, enabling the digital twin to continuously grow, verify, and update.

[0063] In the construction and application of existing digital twin systems for oil and gas fields, a relatively linear process is typically adopted: 3D model construction is completed in the design phase, model display or periodic updates are the main focus during the construction phase, and monitoring and early warning are achieved through IoT data in the operation and maintenance phase. A typical characteristic of this process is that the digital twin model is mostly built once at the beginning of the project and subsequently delivered primarily as a static deliverable. Although some solutions introduce methods such as 4D BIM for progress display, the large amount of change information generated during construction often relies on manual data entry or centralized model revisions at the completion phase, leading to delayed model updates and a high risk of omissions. Simultaneously, change data such as route rerouting, equipment replacement, and parameter adjustments during the construction phase are usually archived in document form, making it difficult to write back to the design model in a structured manner. This results in inconsistencies between the digital twin and the actual entity after completion. Furthermore, real-time monitoring data collected during the operation and maintenance phase is often independent of the construction phase model system, exhibiting significant data fragmentation, making it difficult for the digital twin to form a unified closed loop spanning design, construction, and operation and maintenance.

[0064] To address the issues of static models, difficulty in reverting construction changes, and lifecycle disruptions in the old process, this invention proposes a dynamic growth and verification process for digital twins driven by construction procedures. The core idea is to transform the digital twin construction process from "phased delivery" to "synchronous evolution with construction." In this new process, firstly, entity objects such as oil and gas field engineering equipment and pipeline sections are uniformly encoded using the Entity Object Structure (EBS), ensuring that all subsequent construction deliverables and operational data are uniquely bound to specific objects, thus laying the foundation for lifecycle data organization. Subsequently, this invention abstracts the construction process into a sequence of procedures and generates incremental deliverable data packages for each procedure node. When the completion trigger conditions of a procedure are met, incremental loading operators load the deliverables item by item into the twin state, enabling the digital twin to continuously "grow" with construction activities such as welding, hoisting, and pressure testing, rather than remaining in a static model stage.

[0065] More importantly, this invention introduces a database release mechanism during the construction phase, uniformly recording incremental change data generated during construction and triggering backflow updates through deviation judgment. When a difference is detected between construction data and design data, the system automatically writes the change information back to the design database in a structured manner, thereby ensuring that the design model can continuously reflect the actual state of the construction site. Compared with the old process that relied on centralized data collection and manual model modification during the completion phase, this mechanism significantly reduces the problems of construction change omissions and delivery inconsistencies, enabling the digital twin to maintain a high degree of consistency with the physical entity upon completion.

[0066] Furthermore, during the operational phase, this invention employs a computational-measurement integrated closed-loop verification mechanism to calculate the deviation between the simulation model output and real-time IoT monitoring data. When the error exceeds a threshold, iterative updates of the model parameters are automatically triggered. This allows the digital twin to not only reflect the system status in real time but also adaptively correct model parameters when deviations occur, maintaining high fidelity. Compared to older solutions where monitoring data is only used for display or early warning and lacks closed-loop verification and dynamic correction capabilities, this invention significantly improves the reliability and predictive analysis capabilities of the digital twin.

[0067] In summary, existing technical processes primarily involve static modeling and phased updates, making it difficult to effectively redirect construction changes and resulting in a disconnect between operation and maintenance data and construction data. This leads to deviations and gaps in the digital twin's lifecycle. This invention addresses these issues by adding a dynamic growth mechanism driven by construction procedures, a change redirection mechanism for the published database, and a closed-loop verification and parameter iterative update mechanism based on calculations and measurements. It also eliminates the traditional centralized model modification and manual data entry steps during the completion phase. This allows the digital twin to continuously evolve during the construction phase, consistently update during the change phase, and continuously self-verify and correct during the operation phase, thereby achieving consistent maintenance and reliable delivery throughout the entire lifecycle of oil and gas field engineering projects.

[0068] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Based on the technical essence of the present invention, any simple modifications, equivalent substitutions, and improvements made to the above embodiments within the spirit and principles of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for dynamic growth and verification of digital twins in oil and gas fields driven by construction procedures, characterized in that, Includes the following steps: S1. Input the basic design data of oil and gas field engineering into the digital twin platform, complete the initialization of the digital twin, the splitting and encoding of entity objects based on the entity object decomposition structure, the construction of multi-dimensional sub-models, and output the digital twin base of the oil and gas field. S2. Input the oil and gas field digital twin base and construction procedure planning data into the digital twin platform. Define the construction procedure sequence and configure the incremental data package of deliverables and procedure triggering conditions within the digital twin platform. Output the construction procedure sequence and incremental data package configuration scheme with procedure triggering conditions. The procedure triggering conditions are used to determine whether the construction procedure is completed and trigger the dynamic growth of the digital twin. The procedure triggering conditions include receiving the electronic file generated by the procedure acceptance record, obtaining the online confirmation instruction of the construction visa, receiving the input signal containing material scanning information and installation completion status, or the collected test data meeting the specification requirements. S3. Input the construction process sequence with process triggering conditions, the incremental data package configuration scheme of deliverables, and the process completion data at the construction site into the digital twin platform. Perform incremental mounting within the digital twin platform to achieve dynamic growth of the twin. At the same time, record construction changes and complete the backflow update to the main database of the design stage. Output the updated digital twin and the synchronously updated main database of the design stage. S4. Input the updated digital twin and IoT real-time monitoring data into the digital twin platform, collect the simulation model output data in the digital twin platform and perform calculation and measurement closed-loop verification, output the calculation and measurement deviation results, and output the model parameter iterative update trigger command when the deviation exceeds the preset threshold. S5. Input the model parameter iteration update trigger command and digital twin model parameters into the digital twin platform. The digital twin model parameters include wellbore friction factor, geostress coefficient, flow resistance coefficient, heat transfer coefficient, and equipment performance curve parameters. Perform model parameter iteration updates and automatic verification of the six properties of twin data within the digital twin platform. The automatic verification of the six properties of twin data is achieved by configuring the verification requirements as executable verification rules and associating them with the process trigger conditions. When the process is completed or the digital twin data is updated, the corresponding verification rules are automatically invoked to complete the process. As the construction process progresses, new construction process data and on-site process completion data are cyclically input into the digital twin platform. Steps S2 to S4 are repeated to output a dynamically updated oil and gas field digital twin with a full life cycle.

2. The method for dynamic growth and verification of oil and gas field digital twins driven by construction procedures according to claim 1, characterized in that, In step S1, the digital twin initialization involves constructing the digital twin into a six-tuple that includes a set of oil and gas field engineering entity objects, a set of multi-dimensional sub-models corresponding to the entity objects, a construction procedure sequence, a design phase master database, a construction phase release database, and a set of twin consistency verification functions.

3. The method for dynamic growth and verification of oil and gas field digital twins driven by construction procedures according to claim 1, characterized in that, In step S1, the entity object splitting and encoding based on the entity object decomposition structure is to split the pipe section, valve, structure, and oil and gas field process equipment into the smallest entity units, and generate a unique identity code for each smallest entity unit by processing the object category, spatial location code, equipment tag number or pipeline number through a hash mapping function. The multidimensional sub-model set includes BIM 3D geometric model, GIS geospatial model, flow thermodynamics simulation model, and entity object attribute state relationship data model.

4. The method for dynamic growth and verification of oil and gas field digital twins driven by construction procedures according to claim 1, characterized in that, In step S2, the incremental data package of deliverables is configured in the digital twin platform as a structured data package bound to the unique identity code of the entity object, which includes the process deliverables, the process completion timestamp, and the version number; the process deliverables include the model, parameters, or document deliverables generated by the process.

5. The method for dynamic growth and verification of digital twins of oil and gas fields driven by construction procedures according to claim 1, characterized in that, In step S3, incremental mounting is performed within the digital twin platform. This involves mounting the incremental data package of the results to the digital twin base of the oil and gas field using an incremental mounting operator. The incremental mounting operator can be implemented in one or more of the following ways: attribute appending, parameter overwriting and updating, establishing entity object relationship edges, and generating new model version references.

6. The method for dynamic growth and verification of oil and gas field digital twins driven by construction procedures according to claim 1, characterized in that, In step S3, recording construction changes and completing the backflow update to the main database of the design phase involves first recording the incremental changes generated during the construction phase through the database published during the construction phase. Then, the consistency check function of the twin is used to compare the consistency between the design data and the construction data in terms of geometry, parameters, and relationships to calculate the change deviation. When the change deviation is non-zero, the construction change data is structured and written back to the main database of the design phase.

7. The method for dynamic growth and verification of digital twins of oil and gas fields driven by construction procedures according to claim 6, characterized in that, When version conflicts occur when construction change data is written back to the main database during the design phase, the following strategies are adopted: retaining both versions and marking their priorities, overwriting the design version with the actual construction version, or merging the versions after triggering manual confirmation.

8. The method for dynamic growth and verification of oil and gas field digital twins driven by construction procedures according to claim 1, characterized in that, In step S4, the real-time monitoring data of the Internet of Things is pressure, temperature, flow engineering operation parameters; the calculation and measurement closed loop verification is performed in the digital twin platform, which is to compare and calculate the measurement deviation E of the real-time monitoring data of the Internet of Things and the output data of the simulation model t , the error judgment threshold is set as θ, if E t ≤θ, the verification is passed, if E t >θ, the model parameter iteration update is triggered; the measurement deviation is measured in any one of the Euclidean distance, Mahalanobis distance and relative error function.

9. The method for dynamic growth and verification of digital twins of oil and gas fields driven by construction procedures according to claim 8, characterized in that, In step S5, the model parameter iterative update performed within the digital twin platform is to iteratively update the model parameters by combining the learning rate and the gradient of the loss function. The iteration termination condition is any one of the following: the measurement deviation is less than or equal to a preset threshold, the number of iterations reaches a preset upper limit, or the error change meets the convergence condition. The parameter update algorithm is any one of Kalman filtering, Bayesian update, or least squares estimation.

10. The method for dynamic growth and verification of digital twins of oil and gas fields driven by construction procedures according to claim 1, characterized in that, In step S5, the automatic verification of the six properties of the digital twin data executed within the digital twin platform is achieved by configuring the verification requirements of standardization, completeness, accuracy, consistency, timeliness, and accessibility into executable verification rules and associating them with the process triggering conditions. When the process is completed or the digital twin data is updated, the digital twin platform automatically calls the corresponding verification rules to automatically verify the encoding format, entity objects and their relationships, design data and measured data, database synchronization status, update time and version, data retrieval and access permission status of the digital twin data. If the verification fails, data correction, data synchronization, model parameter updates, or relationship reconstruction are automatically executed according to the anomaly type.

Images