Originally published by Pipeline Technology Journal, February 2024.*
What are the current best practices for managing and preventing corrosion in pipelines?
Rodolphe: Preventing corrosion relies on establishing a set of prevention barriers. While significant progress has been made in the engineering domain (Corrosion coupons, ER Probes, Pigging, etc.), other crucial barriers, such as the human factor and the management system factor, play equally important roles.
Some typical points for consideration include:
- Is the system effectively consolidating corrosion-related data to extract valuable insights?
- Are corrosion and integrity teams collaborating efficiently and effectively?
- Is the sharing and integration of lessons learned effectively taking place within the organization?
These questions underscore the multifaceted nature of corrosion prevention. Therefore, alongside advancements in engineering, the current best practices for managing and preventing corrosion in pipelines include maintaining an efficient management system, competent personnel, and a robust safety culture within the organization. State-of-the-art pipeline integrity management software solutions, such as Synergi Pipeline, provide advanced analytics, such as corrosion growth modelling features, making it easier to monitor, manage and prevent corrosion growth. They also offer improved communication with their capability to seamlessly connect to Corrosion monitoring systems.
How can the uncertainties and challenges associated with complex pipeline geometries, materials, and operating conditions be managed?
Troy: Managing uncertainties and challenges associated with complex gas and oil pipeline geometries, materials, and operating conditions requires a combination of effective monitoring, risk assessment, leak detection, and mitigation strategies. Operators can face a wide range of issues with these situations, including failures such as cracks, leaks, corrosion, and material failure. Remote Telemetry Units (RTUs) can be useful by continuously monitoring for changes in conditions and managing remote automation. Probabilistic risk models should be utilized where uncertainties can influence decision-making and drive continuous improvement in data collection, as they quantify the likelihood and consequences of various scenarios, aiding in informed decision-making. Companies can better prepare for risks by identifying credible threats, assessing their impact, and developing a plan to mitigate them.
Rodolphe: Effectively managing uncertainties requires the implementation of a comprehensive risk assessment, enabling the prioritization of mitigation actions. The usual risk assessment approaches often fail when multiple threats interact (especially if some data is missing), and probabilistic approaches are able to account for data uncertainty but not modeling uncertainties. A new approach is using Bayesian Networks, a form of machine learning, to construct mechanistic and probabilistic risk models. These networks excel in handling both types of uncertainties and offer adaptability in scenarios with missing data. For over a decade, DNV has been using this approach to aggregate information from various sources to address threats like corrosion, stress corrosion cracking, third-party damage, illegal tapping etc. Just take a simple example: an external pipeline corrosion model should not depend only on corrosion but also on other mechanisms such as cathodic protection, coating degradation, vegetation and even weather patterns. This approach allows the creation of truly comprehensive risk models.
In what scenarios is direct assessment preferable to inline inspection for pipeline safety, and what are the trade-offs between these two approaches?
Troy: Industry and regulatory changes often favour ILI over Direct assessment (DA). That said, DA methods can be preferable to inline inspection (ILI) tools in certain scenarios. DA is typically used when ILI is not feasible due to unpiggable pipeline parameters or conditions such as diameter, flow, and other restrictions or difficult to inspect.
The trade-offs between these two approaches are:
- Cost: DA is generally less expensive than ILI and even more so when internal resources are used.
- Accuracy: ILI is more accurate than DA when higher-resolution tools are used
- Time: DA can take longer to complete than ILI when scheduling external vendors and multiple surveys are required.
- Data Volume: DA generates a large amount of data, which can be difficult to manage and analyze.
Rodolphe: While ILI offers unparalleled value by providing vital information about the type, location, and size of anomalies, it falls short in explaining the underlying mechanism causing the defect. This is where the Direct Assessment process becomes invaluable as it addresses the "Why?" question, offering insights into the occurring mechanism and facilitating preventive measures. Ideally, a synergistic approach is recommended where In-Line Inspection and direct assessment complement each other, ensuring a comprehensive evaluation of pipeline safety.
How can data from multiple sources and technologies be integrated to improve the accuracy and reliability of pipeline defect detection and assessment?
Troy: Data from multiple sources and technologies can easily be integrated with proper software tools that support rich analytics from data managed in a standardized format in a common storage repository. Information crucial for pipeline defect integrity assessment is sourced from various channels like pipeline inspection tools, field measurements, sensors, testing, and GIS.
However, the journey of data consolidation introduces challenges, such as issues with diverse formats, alignment variations, and accuracy levels. Extracting meaningful insights from integrated datasets is complex, and ensuring the reliability of integrated data requires robust quality assurance processes. Fortunately, software solutions like DNV’s Synergi Pipeline can serve as a key ally in overcoming these challenges:
- Efficient Integration: The software seamlessly incorporates diverse formats, overcomes alignment variations, and ensures accuracy through standardized protocols.
- Insightful Analysis: Leveraging advanced data analytics techniques becomes streamlined with software, employing machine learning algorithms, statistical analysis, and data visualization tools to extract meaningful insights from integrated datasets. This analytical approach enhances defect detection and assessment accuracy.
- Reliable Data Quality: Software plays a pivotal role in ensuring data reliability through robust quality assurance processes. Procedures such as data cleansing, validation, and verification identify and rectify inconsistencies and inaccuracies within datasets, maintaining high data quality standards.
How can the frequency and scope of inline inspections be optimized to reduce costs and risks?
Troy: The use of in-line inspection is expanding among pipeline operators. The development of new technologies and innovative techniques has helped to improve accuracy, efficiency, and lower costs. The use of software which supports and integrates detailed defect assessment, anomaly lifecycle management, and risk/condition-based inspection scheduling to understand and evaluate mitigation strategies will help reduce costs and risks.
Rodolphe: To optimize inline inspection (ILI) frequency and scope for cost and risk reduction, it's crucial to design a tailored ILI system following API 1163 recommendations for each pipeline's unique challenges. Clearly defining inspection objectives is essential—answering questions like credible mechanisms and expected defects aids in selecting appropriate technologies for valuable results, ultimately leading to cost savings and risk reduction. A comprehensive Threat Assessment is crucial in fine-tuning the ILI system to target and mitigate specific risks, enhancing the efficiency and cost-effectiveness of the inline inspection strategy.
How have recent innovations in pipeline inspection techniques improved the detection and management of potential safety hazards?
Troy: Recent innovations in natural gas and oil pipeline inspection techniques have significantly improved the detection and management of potential safety hazards. Non-destructive testing (NDT) technologies are helping oil and gas producers across all market sectors. Some of the most effective techniques include:
- Ultrasonic Testing (UT): UT instruments facilitate faster setup, reduce inspection time, have none of the health risks associated with radiation, and ensure the full volume of a weld is covered.
- UAVs and Drones: UAVs equipped with sensors are more efficient and cost-effective than traditional methods, simultaneously reducing the risk of human injury.
- Sensors: Optical and other sensors play a pivotal role in detecting leaks, strain, fatigue, and ground movement.
These techniques have enhanced the accuracy, efficiency, cost-effectiveness and safety of pipeline inspections.
Rodolphe: Historically, corrosion was the main focus of the industry's innovations, which explains the current maturity levels of Metal Loss inspection techniques. Now, other damage mechanisms, such as Stress Corrosion Cracking (SCC) and Geohazards, are responsible for many high-consequence failures. As outlined by Troy, the industry has diversified its focus and implemented advanced techniques, with notable advances in:
- Crack Anomalies: The detection of cracks in gas pipelines has substantially advanced with the recent developments in EMAT (Electromagnetic Acoustic Transducer) ILI technology. Ultrasonic tools have also considerably evolved, making these innovations indispensable for the accurate detection, identification, and sizing of crack anomalies in liquid & gas pipelines.
- Material Properties: Challenges like lost pipeline records, non-compliance with as-built records, and changes during the pipeline's lifespan can now be addressed through advancements in ILI tools in identifying hard spots and Pipe Grades, allowing accurate fitness for service of pipelines.
- Pipeline Strain and Movement: Real-time Strain Gauges and High-Resolution ILI Geometry tools provide crucial insights into bending strains and pipeline movement, allowing for proactive measures to mitigate potential risks.
As the industry continues to advance in pipeline inspection, questions arise regarding other critical aspects of pipeline integrity management. Addressing issues such as data errors, overcomplex processes, lack of safety culture, and potential loss of competencies may hold the key to making a more impactful change in reducing pipeline incidents.
What are the key challenges in assessing and mitigating risks associated with pipeline integrity?
Troy: Some of the challenges in assessing and mitigating risks associated with natural gas and oil pipeline integrity include accurate prediction and assessment of internal and external corrosion, construction issues, operational practices, and third-party damage. Many companies also struggle with data quality and the impact that unknown and incorrect data have on risk assessment accuracy.
The US regulatory body, PHMSA, has conducted multiple risk workshops with pipeline operators and has identified general weaknesses in some of the more basic risk models used in the industry. The report provides an overview of methods and tools for improved pipeline risk modelling, including the use of more advanced quantitative and probabilistic models with monetized risk, consideration for data uncertainty, and the identification of mitigative measures.
Rodolphe: Addressing and mitigating risks associated with pipeline integrity involves navigating several significant challenges:
- Tailoring the right model for your system:
Tailoring models to the specific characteristics of the pipeline operators' system and data enhances the accuracy of risk predictions beyond ‘off the shelf’ models.
- Data Quality Issues:
Inconsistent, incomplete, or inaccurate data is a major obstacle to conducting reliable risk assessments. Robust data quality assurance processes, including thorough validation and cleansing procedures, are vital.
- Dealing with Missing Data:
The absence of data often has a significant impact on the final risk results. Employing effective strategies to address missing data, such as probabilistic approaches like DNV's PRA model or tapping into alternative data sources, is crucial for conducting comprehensive risk evaluations.
- Black Box Models:
Black box models, where it is unclear why a certain risk is high, present challenges in identifying targeted risk mitigation strategies. It is crucial to ensure a transparent understanding of risk factors and facilitate effective risk mitigation.
- Understanding Risk Results (Aggregation):
Aggregating risk results necessitates careful consideration, as overlooking nuances in aggregated data may lead to inaccurate risk prioritization. Defining clear Risk acceptance criteria helps avoid oversights in the evaluation process.
In what ways does digital transformation enhance the effectiveness of integrity management?
Rodolphe: In contrast to traditional approaches, digital platforms introduce capabilities that were previously unattainable. They elevate integrity management across multiple dimensions:
- Control: Provide greater control over the integrity of pipelines, empowering operators to stay on top of their priorities.
- Accessibility: Offer a single source of truth, making information and resources easily accessible to all stakeholders, thereby improving transparency and collaboration.
- Efficiency: Streamline processes, speed up commodity tasks and reduce redundant actions, resulting in significant time and cost savings across the organization.
- Intelligence: Harness advanced data analytics to derive valuable insights, enabling data-driven decision-making and predicting future trends and opportunities.
- Communication: Enhance internal and external communication through integrated tools and regular updates, fostering better collaboration and information sharing.
- Judgment: Enable better-informed decision-making by providing stakeholders with comprehensive data and analysis, allowing for more accurate and well-considered judgments.
What are the emerging trends in AI and data analytics that you foresee having a significant impact on pipeline safety and inspection in the near future?
Troy: The rapidly growing technology with AI, Machine Learning, and Data Analytics will quickly transform data-driven decision-making for the utility industry. Here are some of the trends that could have a significant impact on pipeline safety and inspection soon:
- Data-Centric AI - Enhancing model accuracy through improved data quality and larger, richer data sets.
- Generative AI - This technology has the potential to transform big data analytics by generating synthetic datasets and accelerating software development and quality. It opens new avenues for predictive analytics and data visualization.
- Model-driven data analytics – This will guide the development, interpretation, and validation of model algorithms such as Machine Learning for corrosion prediction for unpiggable lines.
Looking forward, what emerging technologies or methodologies do you see as game-changers in the field of pipeline safety and inspection?
Rodolphe: On the technology front, substantial investments are being dedicated to Artificial Intelligence and how it can significantly change the way we manage the integrity of pipelines. I firmly believe that these developments have the potential for a considerable and positive impact on our sector. Some applications include:
- Machine Learning models: These models can consume large amounts of data to provide solutions for problems where conventional methods fall short. Applications range from improving corrosion prediction to optimizing dig selections to predicting the next third-party incident. This enables proactive strategies by predicting potential failures.
- Conversational AI: Introducing conversational AI within the Integrity Management System can significantly enhance communication among all stakeholders of an Integrity Management System. Complex information becomes more accessible, enabling the adoption of new software solutions by reducing users’ resistance to change. This user-friendly interface opens avenues for improved collaboration and understanding among team members, thereby contributing to a more cohesive and informed Integrity Management System.
The Experts
Troy Weyant, Product Manager - Pipeline Product Line, DNV
Troy joined DNV in 1994 and is currently responsible for the pipeline risk & integrity management software strategy and roadmap to meet the needs of the global integrity management market. Before this role, Troy held the position of Principal Integrity Solutions Consultant responsible for the implementation of projects based upon DNV's Asset Integrity Management suite of products. He has also been responsible for the development of DNV's MAOP management solution and has served as a Synergi Pipeline technical lead for large integrity and GIS implementation projects in the US and abroad.
Rodolphe Jamo, Regional Sales Manager - Digital Solutions, DNV
Rodolphe joined DNV in 2022 and has 15 years of experience in the Pipeline Integrity Management field. He holds an MSc in GIS and started his career as a Pipeline Risk Engineer and then moved to Project Management responsible for the implementation of multiple PIM Solutions across the globe. In his role as Senior Key Account Manager, he was also responsible for the provision of tailored integrity and inspection solutions to major gas operators. Rodolphe’s current focus is on supporting operators in their Digital journey via DNV’s suite of Digital Solutions.
Find out more about Synergi Pipeline.
*To view the original article, visit Pipeline Technology Journal