Pipeline digital twin: questions and answers

The system can be implemented in several configurations:

• As a service for preparing reports, technical conclusions, recommendations, and maintenance plans;
• As the DT.pipeline (Digital Twin of Pipeline) software application for modeling the pipeline structure, inputting parameters, performing strength and lifetime calculations, and visualization;
• As a hardware–software complex for evaluating strength and durability, equipped with periodic infrasound diagnostic devices;
• As a real-time leak detection system, using continuously operating infrasound sensors integrated with DT.pipeline analytics;
• As a cloud service, processing field measurements (or virtual data) to assess pipeline strength and residual life.

The full list of required design, operational and diagnostic data is given on a separate page: Workflow and input data. It can also be taken directly from the operator’s documentation.

Yes. The user can independently populate the model with the structural–technological pipeline scheme, infrasound measurement data, and other datasets in predefined templates. Data can also be uploaded automatically via REST API.

• The first priority is the reconstruction of the digital model (schematic) using project, as-built, and operational documentation.
• It is highly beneficial to have digitized data on pipeline materials, foundations, supports, and anchoring points, though this is often unavailable.
• Next, it is crucial to integrate and digitize historical diagnostics and repair data.
• Additional data sources include operational parameters (pressure, temperature) and external loads.

• Yes, the initial inspection is conducted using portable instruments.
• However, continuous monitoring sensors are not intended for strength and durability assessment.
• Permanent sensors are only justified for real-time leak detection functionality.

• The model accuracy is regulated by an engineering safety factor system in cases where information on specific factors is missing.
• The accuracy of coefficients is experimentally derived through testing at various test sites.
• The model calibration is performed based on failure statistics, if available.

• The Digital Twin utilizes a fully open data storage format in tabular form, accessible to any authorized project participant.
• The availability of data and calculation stages can be verified by various engineering specialists.
• Builders and architects are typically not the primary users of the model in standard practice.

• The system typically allows forecasting for the next 1-5 years, with the possibility of adjustments based on new inspection results.
• For most pipeline elements, the remaining service life exceeds 10-25 years and does not require frequent diagnostics.

• It is a pipeline model that not only represents the geometric structure of the construction but also simulates dynamic displacements (vibrations) in six degrees of freedom.
• Each pipeline element is linked to original documentation and datasets, as well as at least 40 reference or calculated parameters.
• An example is available in the online service: pipeline.dtwin.ru.

DATA COLLECTION:

• Infrasound devices record the spatial vibrations of the pipeline in three axes.
• Additionally, mechanical characteristics, welded joints, operational loads, and temperature conditions are considered.

INTEGRAL STIFFNESS ASSESSMENT:

• Changes in pipeline stiffness are analyzed, accounting for defect impact on strength loss.
• Based on dynamic characteristics, deviations indicating a reduction in load-bearing capacity are identified.

FAILURE PREDICTION:

• Material degradation models are used, incorporating detected defects.
• Failure probability is calculated considering accumulated damage and its kinetics.

Thus, the system does not merely detect defects but predicts the loss of operational reliability by linking defect data to residual life calculations.

• Yes, including:
  • Maximum service life and safe operating conditions.
  • Recommendations for maintenance and repairs (M&R).

• The pipeline model contains all necessary element-level information, including:
  • Dates of completed repairs.
  • Repair volumes and details.

• 54,000 km of oil and petroleum product pipelines use the technology to develop operational programs. This has reduced pipeline-related costs by a factor of 2.
• 160,000 km of gas pipelines have also integrated this technology to optimize operations and enhance reliability.

The DT.pipeline applies a combination of classical and modern engineering methods:

• Digital signal processing algorithms — Laplace transforms, spectral analysis (B. Gold, C. Rader, R. Lyons, L.M. Goldenberg, A.B. Sergienko, etc.);
• Beam-on-elastic-foundation models adapted for pipelines (A.N. Krylov, V.I. Feodosiev, Yu.N. Rabotnov, etc.);
• Stress–strain state (НДС) analysis with local stress concentration effects (Feodosiev, Rabotnov, Bauman MSTU, E.O. Paton Institute, etc.);
• The method of local stresses — an engineering theory of strength and fatigue life.

The DT.pipeline system uses fluid parameters — pressure, temperature, density, viscosity, and velocity — as input data for load and durability calculations.

1. Pressure: both constant and variable components (operating, pulsation, surge) are modeled as functions of time and spatial coordinate, influencing axial and hoop stresses.
2. Temperature: affects thermal elongation and elasticity modulus; thermal gradients along the pipeline and differences between aboveground/underground sections are considered.
3. Velocity and viscosity: contribute to dynamic load estimation through generalized hydrodynamic pressure terms.
4. Mass and density: the transported medium’s mass adds to bending and support loads in the combined stress–strain state model.
5. Non-stationary regimes: during startups, shutdowns, or surges, measured vibration and pressure data from IFC.1 sensors are used to build load time series for fatigue assessment via Miner’s rule.
6. Integrated model: all flow parameters are aggregated into a vector of operational loads, passed into the DT.pipeline solver, producing detailed stress, strain, and remaining-life distributions for every element.

1. Data input: DT.pipeline does not directly measure chemical composition; parameters (H₂S, chlorides, salts, humidity) are imported from lab or SCADA/LIMS systems.
2. Model use: these parameters adjust corrosion-impact coefficients and wall-thinning rates within the the industry methodology for assessing the strength and service life of pipes and welded joints with defectsframework.
3. Limitations: DT.pipeline does not simulate corrosion kinetics or electrochemical reactions — it uses these factors as corrective multipliers in the mechanical model for conservative lifetime estimation.

Verification is carried out at six levels:

1. Geometry and connectivity of the model are checked against as‑built documentation.
2. Comparison of calculated and measured amplitude–frequency characteristics (difference ≤10–15%).
3. Correlation of high‑stress zones with known defect locations.
4. Comparison of predicted remaining life with results of follow‑up inspections.
5. Certification against industry standard the industry methodology for assessing the strength and service life of pipes and welded joints with defects.
6. Certification of the mathematical model and calculation software in the SEC NRS (Scientific and Engineering Centre for Nuclear and Radiation Safety): software attestation certificate No. 673 of 09 July 2026, issued by the Expert Council for Software Attestation under Rostechnadzor.

• Proprietary technologies, experimental data libraries, and strength and durability assessment methods are utilized.
• Algorithms for digital signal processing are applied, including:
• Laplace transformation
• Spectral analysis methods (B. Gold, C. Rader, R. Lyons, L.M. Goldenberg, A.B. Sergienko, etc.).
• Beam analysis algorithms on elastic foundations adapted for pipelines, developed by:
• A.N. Krylov
• V.I. Feodosiev
• Y.N. Rabotnov, etc.
• Both classical and modern stress-strain state analysis methods are used, considering stress concentration effects:
• V.I. Feodosiev
• Y.N. Rabotnov
• Research from the welding technology departments of LPI, Bauman MSTU, ChPI, E.O. Paton Institute, etc.
• The local stress method – an engineering theory of strength and durability – is implemented.

• Materials available for inspection include:
• Polypropylene: PP-R PN20, PP-R PN25
• Concrete
• Brass: Лц40Сд
• Steel:
• St3
• Steel 20
• Steel 20 (200°C)
• 08Х18Н10Т
• A403 WP321-WX
• A358 GR.321C1
• A182.GRF321
• A182 F321
• A312 TP321
• 13ХФА-K52
• 10Г2ФБЮ
• 09Г2С
• Cast Iron
• The material reference database can be expanded.

• External loads are accounted for comprehensively, including:
  • Temperature effects
  • Wind loads
  • Snow loads
  • Soil and other environmental impacts.

• The condition of foundations and structures is assessed as a set of quantitative parameters that define the object as a technical system.
• The primary objective is to identify key parameters affecting reliability and determine the most critical ones.
• Key parameter groups include:
• Foundation structure (dimensions, shape)
• Mechanical properties of materials
• Defects and damage
• Loads and external impacts
• A probabilistic approach is applied to account for the combined influence of heterogeneous factors on strength and reliability.
• All physical values contain uncertainties, which include:
• Incomplete information due to limited observations
• Measurement errors
• Assumptions in computational models
• To reduce uncertainty, additional data sources and Bayesian statistics are utilized.

• The strength and durability of expansion joints are determined by the number of completed cycles under:
  • Proper operating conditions.
  • Compliance with maintenance and repair regulations.
  • Displacements within permissible limits.

• If no additional environmental data is available, the system assumes the worst-case scenario for corrosion progression.
• This means that the estimated corrosion rate is determined based on the maximum possible impact of external factors.

Key conditions affecting corrosion acceleration:

• Electrical power cable crossings – potential influence of stray currents.
• Water body crossings – high humidity and possible presence of aggressive environments.
• Zones with unstable soil conditions – increased mechanical stresses.
• Technological areas exposed to chemical reagents – potential for accelerated metal degradation.

If additional data is available, the system adjusts the corrosion rate based on:

• Local corrosion measurements and pipeline diagnostics.
• Information on pipeline protection (insulation, cathodic protection).
• Operational experience from similar sections.

• The device measures vibrations only at control points, so there are no theoretical limitations on pipeline length.
• All data is reconstructed through mathematical modeling.
• Estimated device coverage ranges:
• Diagnostic mode: up to 50 km.
• Real-time leak detection mode: from 5 to 50 km (depending on pressure).

• The pipeline mathematical model is reconstructed using schematics, field measurements, and material data.
• The absence of specific data does not prevent model creation, but in such cases:
• More conservative tolerances will be applied.
• Safety margins will be calculated with additional engineering reserves.

• Assessment and field measurements can be performed on pipelines of any age.
• There are documented cases of 100-year-old pipelines being successfully monitored.

We always operate under data limitations.

Some data shortages are compensated by:

• Field instrumental measurements
• Linear measurements
• On-site pipeline inspections
• Applying increased tolerances and safety margins in technical condition assessments

Two instrument types are used:

• MOD1 (autonomous diagnostic) — portable device with internal power supply and data logging (up to 72 min sessions, 500–1000 Hz, temperature –15…+50 °C);
• MOD2 (networked “desert” type) — continuously powered 24 V device for real-time measurements.

Both support pipe diameters 13.5–1420 mm, explosion-proof class 2Ex ic IIA T6 Gc X, IP54 protection, M8 or clamp mounting, and accuracy ±0.05 g / ±0.1 dps / ±0.05 G.

The instrument measures linear and angular displacements of the pipeline using 3-axis accelerometers and gyroscopes, allowing reconstruction of vibration modes, stress–strain dynamics, and indirect detection of anomalies caused by corrosion, weakening, or defects.

Calibration (tariration) is carried out on a specialized calibration bench that provides six controlled orientations (3 accelerometer axes and 3 gyroscope axes) with known accelerations and angular velocities.

Sessions are executed in six directions (“forward”, “backward”, “up”, “down”, “left”, “right”); results form correction matrices minimizing sensor errors.

Calibration is required annually or after transportation/repair, and results are logged in a calibration protocol with timestamps and operator signature.

Sampling frequency: accelerometer/gyroscope — 500–1000 Hz; magnetometer — 10 Hz.

Session duration: MOD1 up to 72 minutes per session (autonomous), MOD2 continuous real-time transmission.

Accuracy: ±0.05 g, ±0.1 dps, ±0.05 G; time synchronization ±1 ms; total vibration amplitude/frequency error ≤3%.

The procedure includes:

1. Inspection of casing integrity, labeling, and power supply (24 V for MOD2).
2. Sensor self‑testing (messages “Test FAIL/OK”).
3. Calibration of accelerometer and gyroscope zero offsets.
4. Noise floor and timing synchronization check.
5. Orientation fixation during installation and metadata registration.

A dedicated calibration bench and procedure have been developed. Calibration is required at least once per year.

MOD1 (autonomous): records to microSD, duration up to 72 minutes, sampling rates 500–1000 Hz, data transfer via Wi‑Fi or cable.

MOD2 (networked): continuous data transmission over RS422‑Ethernet, 24 V power supply.

Collected data include accelerometer, gyroscope, magnetometer, and temperature time series, along with service events and timestamps.

Applicable pipe diameters: 13.5–1420 mm;

Operating temperature: −15 °C to +400 °C;

Explosion protection: 2Ex ic IIA T6 Gc X;

Ingress protection: IP54.

Mounting: magnetic base and/or M8 clamps and studs (tightening torque ≤8 N·m).

Power supply: 24 V; autonomy up to 24 hours (MOD1).

• The devices are installed on exposed pipeline sections (on rigidly fixed supports) to collect readings at control points.
• Underground installation is not required if there are no designated service points.
• Optimal control points are chosen at pipeline branching locations, as these areas are typically accessible for maintenance (e.g., vaults, exposed sections).

Infrasound monitoring devices used in the system include the IFC.1 in MOD1 mode (autonomous).

Technical specifications of the IFC.1 device:

• PURPOSE: Monitoring and assessment of the technical condition of pipelines, buildings, and structures.
• SENSORS:
• Three-axis accelerometer (range: ±2 g, frequency: 1000 Hz).
• Three-axis gyroscope (range: ±250 dps, frequency: 1000 Hz).
• Three-axis magnetometer (range: ±4 gauss, frequency: 10 Hz).
• Temperature sensors (range: -40 to +85°C, frequency: 1 Hz).
• DATA RECORDING:
• Built-in 32 GB non-volatile memory (MicroSD).
• Data transmission via Wi-Fi or wired connection.
• POWER SUPPLY:
• Lithium iron phosphate (LiFePO4) batteries, capacity 3200 mAh.
• Autonomous operation for at least 24 hours.
• OPERATING CONDITIONS:
• Pipe outer diameter range: 13.5 mm – 1420 mm.
• Operating temperature: -15 to +50°C.
• Explosion protection rating: 2Ex ic IIA T6 Gc X.
• Dust and moisture protection level: IP54.
• ADDITIONAL FEATURES:
• Operating modes:
• Mode 1: Testing, calibration, programming.
• Mode 2: Autonomous diagnostics and data recording.
• Mode 3: Sleep mode and data transmission to an external terminal.

Interpretation of the collected data allows for calculating pipeline loads and predicting remaining operational life.

• Yes.

• Yes, for buildings and structures, including bridges.

• The device undergoes calibration once every 3 years.

The inspection was conducted under normal operating conditions as per the regulatory norms for the given facility.

The calculations accounted for possible variations in process parameters up to the maximum allowable limits.

For inspection and subsequent calculations, at least two sensors must be installed on the pipeline, preferably near the terminal zones.

Two devices allow for:

• Separation of overall pipeline vibrations from its natural oscillations
• Determination of the pipeline’s dynamic characteristics

For branched pipelines, sensors should be placed:

• At terminal zones (on shut-off valves)
• At branching points (flanged tees)

The placement of devices is also determined by the following factors:

• Sensors are installed only on uninsulated sections (except for painted surfaces)
• The installation location must be accessible without restricting the operating mode

Defect ranking can be performed by any quantitative attribute calculated in dynamic analysis — stress level, accumulated damage, or safe operation period.

In practice, elements are ranked by residual life or risk index, combining defect severity and potential economic damage. Noise is reduced by statistical averaging and spectral filtering of sensor signals.

1. Chemical aggressiveness: the system does not model chemical corrosion (H₂S, chlorides, salts) — these are user inputs.
2. Internal flaws: instruments detect only external dynamic manifestations; subsurface microdefects without mechanical response are not captured.
3. Extreme conditions: frequencies >1000 Hz or temperatures >+400 °C exceed sensor and model accuracy limits.

Elements are ranked according to safe operation period (remaining life):

1. Group I — remaining life shorter than time until next inspection (unacceptable risk, immediate repair required).
2. Group II — remaining life shorter than inspection interval considering repair time (repair before deadline required).
3. Group III — remaining life exceeds inspection interval (repair not required until next scheduled inspection).

The defect card includes:

• Identifiers (pipeline_id, element_id, defect_id);
• Defect type (geometric, corrosion, weld, crack);
• Parameters (depth, length, width, coordinates);
• Detection date and inspection method;
• Impact evaluation on durability (D, σeq);
• Recommendations and execution status.

Data exchange formats — JSON or CSV.

• The Pipeline Digital Twin utilizes a leak detection method based on analyzing the dynamic characteristics of the pipeline itself, rather than just the parameters of the transported product. The method is based on:
  • Dynamic analysis of mechanical loads – modeling pipeline behavior during a leak, considering induced oscillations and their amplitude-frequency characteristics.
  • Infrasound sensors – installing measurement devices at intervals depending on pipeline pressure, detecting pipeline vibration deviations from the norm.
  • Frequency response module – pre-calculating frequency patterns, which are updated during calibration.
  • Integration of thermohydraulic and mechanical dynamic calculations – ensuring more precise identification of leak zones based on characteristic structural dynamic deviations.
  • How the Digital Twin method differs from common leak detection methods:
  • Standard Leak Detection Systems (LDS) primarily rely on flow parameter analysis (mass balance, negative pressure wave method, RTTM), which limits sensitivity (sensitivity and detection time are limited by the flow-measurement principle itself) and results in detection times of around 30 minutes.
  • The Digital Twin method allows detection of leaks as small as 0.1% by registering mechanical deformations in the pipeline and analyzing vibration characteristics.
  • Traditional methods have a high probability of false alarms, especially during pressure fluctuations, pump startups/shutdowns, and other process changes. The Digital Twin method minimizes false alarms by analyzing pipeline behavior instead of fluctuating flow parameters.
  • Thus, the Digital Twin technology provides more accurate and faster leak detection, reducing false alarms through a comprehensive approach that combines dynamic modeling and localized pipeline structure monitoring.

Based on the results obtained at major Russian pipeline and gas transmission operators, the accuracy of the methodology is sufficient to reduce failure rates and unit operating costs by half.

The pipeline reliability assessment methodology is based on rigorous physical and mathematical models and considers:

• Structural parameters, environmental factors, internal and external loads.
• Characteristics of pipes, welds, defects, and their variations.
• Engineering safety margins.

The methodology incorporates:

• Experimental data on the mechanical properties of pipes and welds.
• Results from in-line and external diagnostics.
• Statistical methods for estimating failure probabilities.
• Modeling of damage kinetics and defect growth.

THE RELIABILITY OF THE OBTAINED DIAGNOSES IS ENSURED BY:

• Statistical processing of experimental and diagnostic data.
• Accounting for uncertainties (parameter variations, diagnostic errors).
• Risk modeling based on reliability theory.

Diagnosis accuracy is adjusted using an engineering safety margin, which depends on:

• Measurement precision.
• Relevance of diagnostic data.
• Correctness of defect evaluation.

The system enables optimal planning of future work volumes, including:

• Repair of supports and hangers.
• Replacement of pipeline sections and bends.
• Reworking of defective weld joints.

PLANNING IS BASED ON ANALYSIS OF:

• The current technical condition of the pipeline (strength and durability).
• Characteristics of weld joints.
• Identified defects and their progression.

The forecasting horizon depends on:

• Availability of diagnostic data.
• Rate of defect development.
• Operating conditions.

The methodology assesses component longevity, allowing projections beyond 30-50 years.

Recommendations are generated automatically based on complex stress–strain and residual-life analyses, comparing actual values with allowable thresholds.

Each element is classified into risk groups (A/B/C). Results are provided in tabular and report formats or as structured JSON objects for ERP/maintenance systems.

Detailing reaches element level (centimeter precision), including specific defects and proposed maintenance actions.

Reports: generated via R Markdown/LaTeX in PDF/HTML/DOCX, plus CSV/JSON exports.

Sections include remaining-life summaries, risk groups, stress maps, defect tables, and repair schedules.

Dashboards: interactive 3D–5D pipeline models (geometry + load + defects + timelines), thermal maps, frequency spectra, and scenario simulations.

Integration: REST API (JSON), batch CSV/XML imports to SAP PM, 1C TOiR, Maximo; OPC UA/Modbus data tags for SCADA/MES; internal web dashboards and secure file exchange.

A continuous calculation pipeline operates in the system: recalculations run automatically when new measurements or updated parameters are received.

Trigger conditions: new measurements, structural-scheme changes, or operator request.

Regular updates may also be scheduled (monthly, per inspection cycle); recalculated values and reports are versioned with full change history.

The report contains:

1. General information about the facility;
2. List of input data;
3. Calculation scheme and load parameters;
4. Maps of displacements, stresses, and strains;
5. Table of remaining life and safe operation periods;
6. Recommendations for repairs and corrective actions.

The update procedure includes:

• Annual recalculation based on new measurements and operational data;
• Model versioning with timestamp and author tracking;
• Retention of previous versions for dynamic comparison of stress–strain and damage evolution;
• Automatic data consistency check when new measurements are uploaded.

• Yes, the operational program includes recommendations for the next diagnostic schedule.
• For 85% of pipeline sections/elements, the next diagnostic interval will be over 5 years.
• For 12% of elements, the next diagnostic interval is approximately 1 year.

Based on the results of calculations, an optimal operation program is necessarily formed, including:

• Diagnostics and condition monitoring
• Current and major repairs
• Modernization and reconstruction
• Comprehensive protective measures

In the final conclusion section, excessive information is not included.

The Main Stages section of the report details all performed work and results that can be obtained using this method.

The scope of work included only selected tasks, and comparison with project documentation was not part of the technical assignment.

Based on infrasound control results, certain pipeline elements were found to have stress levels exceeding the permissible design norms.

Potential causes of excessive stress include:

1. Project non-compliance with construction norms
2. Deviations from design solutions during construction
3. Changes in pipeline elevation over time
4. Deviations from permissible operating modes
5. Excessive vibration from pump units

Incoming data: time series of pressure, temperature, flow, valve positions, alarm signals, and process modes (startups, shutdowns, pumping cycles) in JSON via REST API.

Processed data: derived dynamic loads, stress–strain states, fatigue damage, and remaining-life forecasts.

Output data: aggregated risk indices, limit-exceedance zones, and repair recommendations exported back to SCADA/MES or maintenance systems in JSON/CSV for automatic update of operational dashboards.

Neural networks are not used as black boxes but as auxiliary optimization tools to fine-tune physical models — adjusting coefficients, boundary conditions, and material parameters using real measurements.

They are applied in the adaptive calibration block, trained on historical measurement and inspection data to find stable correlations between defects, loads, and vibration responses.

All AI outputs are verified by analytical recalculation and comparison with physical inspection data, ensuring engineering interpretability and compliance with the industry methodology for assessing the strength and service life of pipes and welded joints with defects.

Yes. DT.pipeline can be fully deployed within the customer’s perimeter, with all data stored and processed locally.

The pipeline digital twin information model is structured hierarchically:

• Level 1 — Pipeline system;
• Level 2 — Pipeline section;
• Level 3 — Elements (pipes, elbows, tees, valves, welded joints, supports).

Each element is defined by geometric parameters (3D coordinates, diameters, wall thickness, orientation, supports), material properties (grade, strength characteristics, low-cycle fatigue parameters), and operating conditions (pressure, temperature, cyclic loading, fluid mass).

Calculation results include distributions of loads, moments, stresses, strains, accumulated damage, and remaining life.

Data exchange is supported in JSON, XML, and CSV formats (JSON is default). A typical structure includes element identifiers, geometry, materials, loading parameters, stress–strain results, and residual life indicators.

Integration can be performed via REST‑API.

Flow 1: SCADA/MES ↔ IOT.SW.1 (software of the DT.Pipeline infrasound measurement device) — exchange of current condition parameters (pressure, temperature, vibration, damage indices).

Flow 2: ERP/Maintenance ↔ DT.Pipeline — exchange of maintenance plans, defect cards, and inspection results.

Data are transmitted in JSON format with OAuth2 authorization and TLS 1.2+ encryption.

Each data tag includes element identifiers (pipeline_id, element_id), measurement date, parameter code, value, and units.

DT.Pipeline — Digital Twin of the Pipeline (hardware‑software complex for leak prevention and detection).

The system architecture is segmented into zones:

• Field devices and gateways (isolated LAN network);
• Calculation and data storage servers (enterprise internal network);
• Integration gateway (DMZ).

Data are transmitted via secure channels (TLS); storage uses encrypted volumes; role‑based access control is implemented; all actions are logged; model versions and security events are recorded and audited.

Base REST methods:

• GET /models/{id} — retrieve the digital model with parameters;
• POST /measurements — upload measurement data;
• POST /calculations — initiate a calculation process;
• GET /results/{calc_id} — retrieve results;
• GET /recommendations — obtain the list of recommended actions.

Data format — JSON, SI units, timestamps — ISO8601. Responses include status, error codes, and messages.

• Local deployment: server and database located within the enterprise; permanently installed devices are connected via LAN; portable/diagnostic devices store data internally (microSD) and are periodically offloaded via USB or card removal.
• Hybrid deployment: local computations with a cloud‑based reporting dashboard.

Minimum requirements: server — 8 cores, 16 GB RAM, 1 TB storage; workstations — x86 CPU, 8 GB RAM, Windows/Linux OS.

The pipeline.dtwin.ru system is an online platform for creating, storing, and analyzing data on the technical condition of pipelines.

To ensure data security, the following modern protection methods are implemented:

• Access control and user privilege management
• Multi-level authentication and account protection
• Data encryption during transmission and storage
• User activity logging to track changes
• Automated data backup system
• The security architecture follows ISO/IEC 27001 controls and NIST recommendations; no certification against these standards has been carried out

Additionally, monitoring and cyberattack protection mechanisms include:

• Intrusion Detection and Prevention Systems (IDS/IPS)
• Antivirus protection and vulnerability analysis
• Regular security audits

• The proprietary software 'Pipeline Digital Twin' is used, officially registered in the Russian software registry.

Calibration applies to all sensitive elements — accelerometers, gyroscopes, magnetometers, and temperature sensors — checking zero drift, sensitivity coefficients, and linearity (±2 g, ±15.6 dps, ±4 G, –15…+50 °C).

Software–hardware checks include time sync accuracy (≤1 ms), sampling stability (500–1000 Hz), and correct calibration-matrix generation and data transmission integrity (USB/RS-422/Ethernet).

The equivalent classifications under international economic standards would include:

• Research and development in engineering and applied sciences.
• Risk management and safety.
• Software development for industrial and engineering applications.

• The technical condition assessment is conducted in accordance with modern pipeline design and operational standards.
• The methodology includes dynamic elasticity theory analysis, strength calculations, and residual life assessment.
• Applicable regulatory documents:
• Domestic standards: SNiP, SP, PNAE, GOST 32388-2013
• International standards: ASME-B31.1 Power Piping
• Personnel performing inspections are certified according to qualification requirements for non-destructive testing specialists.
• Currently, regulatory documents for infrasound control are under development, and it is advisable to plan for the creation of organizational standards to formalize this inspection method.

• Currently, there are no national regulations for calculating the residual service life of pipelines.
• In this study, the residual life assessment was performed using a major Russian pipeline operator’s corporate standard:
• the industry methodology for assessing the strength and service life of pipes and welded joints with defectsDetermination of Strength and Durability of Pipes and Welded Joints with Defects.
• The methodology has been adapted for industrial pipelines inspected using infrasound control.
• For the most pipeline operators has no official standards exist for evaluating the longevity of industrial pipelines.
• The presented calculations in the reports are advisory in nature, as this approach is being used for the first time.
• In the long term, it is advisable to develop and approve corporate standards:
• Methodology for Determining Strength and Durability of Objects
• Infrasound Inspection of Objects

The application of the method extends beyond pipelines. With necessary adaptations, it can be used for:

• Buildings and structures (tanks, bridge structures, trestles, foundations, etc.)
• Process pipelines, fittings, and shut-off valves

The method cannot be applied without significant modifications to:

• Machines and engines
• Rotating mechanisms

Currently, there are no established regulatory documents (NDT) governing field survey methods, computational approaches, device specifications, or certification requirements.

Devices must be installed on pipeline sections free of insulation, excluding paint coatings.

The presence of insulation outside the sensor installation area does not affect measurement accuracy.

Infrasound noise is recorded by the devices as one of the external influences.

The residual life assessment methodology has been applied to over 75,000 km of main oil and petroleum product pipelines (including process pipelines) at a major Russian pipeline operator since 2005 and remains valid to this day.

Key aspects of the methodology:

• There are no universal federal regulations for pipeline residual life calculations.
• To address this, an internal regulatory document (NDT) was approved within the organization to standardize the calculation methodology.
• The reports indicate that the methodology has been adapted for process pipelines and is based on proven operational practices.

What is proposed for thec customeer?

• Development and approval of a corporate NDT to regulate the residual life assessment of process pipelines.
• The current calculation for the customer is a pilot project and serves as a recommendation only.

This method is not a vibration-based method and is not limited to measuring vibrations at discrete points.

Why is GOST 32569-2013 not applicable?

• This GOST does not regulate methods for calculating actual loads and stresses along the entire length of a pipeline, including vibration-induced loads.
• It does not account for material properties of the pipeline, including their changes during operation.
• Vibration-based control only detects vibration levels at measured points but does not determine the overall stress-strain state of the pipeline.
• Pipeline failures typically occur at unmeasured points, making it impossible to detect them directly using vibration control methods.

Limitations of GOST 32569-2013:

• Only regulates defect rejection based on four vibration levels (Level 4 – immediate correction required).
• Does not characterize the operability of individual pipeline elements or the pipeline as a whole.
• Does not provide reliability assessments, particularly regarding service life predictions.

Unlike vibration-based control, the infrasound analysis method:

• Assesses loads and stresses along the entire pipeline length.
• Enables residual service life and reliability calculations for pipeline elements.

Experts followed the methodology for assessing technical condition, strength, and durability, which has been applied at major Russian pipeline and gas transmission operators for over 15 years.

Why is this method not a vibration diagnostic method?

• It is an integrated method, capable of processing results from various technical diagnostic techniques, including vibration monitoring.
• GOST 32569-2013 regulates only vibration levels but:
• Does not describe methods for calculating loads and stresses along the entire pipeline length.
• Does not account for material properties and their changes over time.
• Does not allow for an assessment of pipeline longevity and reliability as a whole.
• Vibration monitoring detects vibration levels only at measured points, whereas pipeline failures typically occur at unmeasured locations, making them undetectable by conventional vibration diagnostic methods.

Why were proprietary instruments used?

• Experts used proprietary monitoring devices due to the absence of diagnostic data provided by the Client.
• The integrated analysis method allowed for:
• Calculation of loads, stresses, and residual service life for each pipeline section.
• System-level pipeline analysis, rather than evaluating only discrete measurement points.

• This method is not a vibration diagnostic method, but an integrated approach that combines various technical diagnostics, including vibration analysis.
• Currently, there are no universal federal standards for evaluating structural strength and durability using infrasound control.
• The methodology is based on proven practices and standards used at major Russian pipeline and gas transmission operators.
• Personnel certification follows the qualification requirements for non-destructive testing and pipeline diagnostics specialists.

If the Client is interested in implementing this method, it is advisable to develop and approve corporate standards, including:

• “Methodology for Determining the Strength and Durability of Objects”
• “Infrasound Monitoring of Objects”

These documents should regulate:

• Methods for calculating strength, durability, and reliability indicators
• Technical requirements for infrasound monitoring devices
• Certification and licensing procedures for their application at the customer’s facilities.

Technical Condition Monitoring in the Pipeline Digital Twin is a comprehensive assessment of pipeline durability and reliability. Unlike traditional diagnostics, it is based on an integrated analysis of the entire pipeline, rather than individual components.

Key differences:

1. Operational and failure history
   • Technical Condition Monitoring considers the full history of construction and operation, analyzing failure and damage statistics.
   • Standard diagnostics does not account for failure statistics.
2. Diagnostic methods
   • Technical Condition Monitoring applies infrasound monitoring of the entire pipeline, as well as dynamic and static load calculations.
   • Standard diagnostics uses localized methods, such as defectoscopy of individual pipes or welds and ultrasonic thickness measurements.
3. Identifying defect causes and mechanisms
   • Technical Condition Monitoring analyzes all diagnostic data throughout the pipeline’s service life, including weld inspections from construction.
   • Standard diagnostics does not account for material properties or actual operating loads.
4. Defect progression forecasting
   • Technical Condition Monitoring calculates durability and residual service life indicators for each pipeline element.
   • Standard diagnostics only provides estimated defect growth rates for corrosion and erosion, while all other defects are assessed using a pass/fail method.
5. Maintenance planning
   • Technical Condition Monitoring generates recommendations based on calculations, including:
   • Additional defectoscopic inspections
   • Operating mode restrictions
   • Repairs (insulation and/or pipe replacement)
   • Adjustment of pipeline elevation and alignment
   • Engineering protection in geologically complex areas
   • Standard diagnostics only generates recommendations based on detected defects.

1. Failure cost estimation: expected damage from leaks or failures = probability × cost of consequences (repairs, downtime, product loss).
2. Maintenance optimization: repairs are scheduled at the point of minimal combined cost of risk and maintenance (min(C_repair + E_risk)), preventing unnecessary interventions.
3. Economic efficiency: compares calendar-based vs risk-based maintenance, quantifying cost reduction, extended lifetime, and reduced downtime; the effect is expressed as annual savings and lower total cost of ownership (TCO).

Pilot applications conducted on assets of major Russian pipeline and gas transmission operators, and industrial facilities.

Results: reduced failure rates, a twofold reduction in specific maintenance costs, and a threefold decrease in physical repair volumes.

High‑stress zones were identified; safe operation periods refined; repair planning accuracy and diagnostic targeting improved.

The process includes:

1. Collection and analysis of documentation;
2. Development of the digital model;
3. Execution of infrasound measurements using DT.Pipeline instruments;
4. Calculation of stress–strain state and residual life;
5. Report generation with recommendations;
6. Data transfer to the customer and personnel training;
7. If required, system refinement for real‑time leak detection functionality.

• Lower probability of failures and accidents.
• Reduction in failure intensity by up to a factor of two.
• Reduction in unit operational (reconstruction) costs of pipelines by up to a factor of two.

• At facilities where the technology has been applied, failure rates and cost of ownership were reduced. The size of the effect depends on the initial condition of the asset and on the completeness of the input data, and is assessed case by case.

The cost depends on the length and diameter of the section, the input data available and the scope of work. We prepare a quotation for the specific facility — describe the task and we will put one together.

The total time costs for the full scope of work include:

The total workload for inspecting and analyzing 457 meters of pipelines was 98 man-days.

4 specialists were involved in the work.

For 100 meters of industrial pipelines, the average workload is 21 working days (per specialist), including:

• 3 working days – for inspection
• 18 working days – for calculations, report preparation, and conclusions

DETAILED BREAKDOWN OF WORKLOAD:

• Inspection (infrasound monitoring): 14 man-days
• Preliminary analysis – 2 man-days
• Direct measurements – 6 man-days
• Digitalization of measurement results – 3 man-days
• Frequency response calculations – 3 man-days
• Calculation and analysis: 64 man-days
• 3D model development – 24 man-days
• Dynamic analysis – 24 man-days
• Stress-strain state calculations – 8 man-days
• Durability assessment – 8 man-days
• Report preparation and recommendations: 16 man-days
• Documentation of calculation results – 12 man-days
• Development of conclusions and recommendations – 4 man-days
• Final conclusions: 4 man-days