•
Sep 24
Sand Technologies
It is essential to understand what is meant by mass balance in the oil and gas industry and why it is so important. Mass balancing refers to the process of accounting for the movement of raw materials, intermediate products and finished products, to ensure that the total mass entering a system equals the total mass leaving the system. The differences in totals should be quantified, such as leakage, inefficiencies and chemical reaction losses, among others. It is more than just simple math, and in part, this is why mass balance is a challenging concept.
The process is crucial for several reasons: it facilitates accurate inventory management, optimizes process efficiency, ensures regulatory compliance and provides reliable data for financial reporting and inventory management. Mass balance enables companies to monitor their operations, enhance profitability and minimize environmental impact.
Identifying potential oil and gas reserves
Extracting resources from the earth
Preparing resources for transportation
Bringing oil and gas to the surface
Moving resources to storage and refining
Holding resources for
future use
Creating chemical products from oil and gas
Transforming resources into usable products
Selling products to
consumers
Using products in various sectors
Identifying potential oil and gas reserves
Extracting resources from the earth
Preparing resources for transportation
Bringing oil and gas to the surface
Moving resources to storage and refining
Holding resources for
future use
Creating chemical products from oil and gas
Transforming resources into usable products
Selling products to
consumers
Using products in various sectors
When considering mass balance, there are multiple lenses with different expectations (time/quality) to assess a process. There is a “process” lens, typically at the unit level, where raw products transform into finished products.
Additionally, there is a “product” lens, measuring how much of a finished product ties to customer sales, movements and accounting. Accuracy requires an understanding of each process. Furthermore, the process level is typically measured daily, while the product level is measured weekly or monthly.
Adding to the complexity is the human side or “copilot” concept. The best mass balance within Refining occurs when a hydrocarbon accountant works with the process engineer to incorporate operational upsets or other events that might impact mass balance. These events could include flaring, taking an off-spec product, and re-circulating/re-processing it to bring it on-spec, or even blending products to meet specification.
When product accounting was initially centralized and later offshored, the copilot concept deteriorated, even though the technology remained essentially unchanged, resulting in a less accurate outcome. Companies have invested tens of millions into automating mass balance in oil accounting systems for refining, as well as similar solutions across the entire molecule life cycle. It is essential to recognize that automation and AI alone (even if key enablers) may not deliver the full value, without a human in the loop to refine model calculations.
Provides a clear picture of materials entering and leaving
Detects leaks and faulty equipment early
Assesses and minimizes environmental effects
Tracks changes affecting financial accounting and profitability
Helps identify issues and improve operations
Ensures adherence to material use and emissions regulations
Incorporates sustainable production and promotes circularity
Achieving end-to-end mass balance is challenging, as numerous data sources must be collected and integrated across various stages of the oil and gas value chain. Modern facilities are inundated with data, emanating from a vast number of sensors, measurement points and diverse data streams. This data often arrives in heterogeneous formats and from various sources, including flow meters, tank level indicators and laboratory analyses, making consolidation a gargantuan task. Measurement inaccuracies and biases further compound the complexity of the data.
| Data Sources for End-to-End Mass Balance | ||||
| Exploration | Upstream | Downstream | ||
| Well logs and core data | Inlet and outlet flow rates | Receipts from various transportation methods (tankers, pipelines) | ||
| Production reports and databases | Fluid density measurements | Process unit input and output measurements | ||
| Laboratory analysis of fluid samples | Line pack mass calculations (change in pipeline inventory) | Tank inventory data | ||
| Seismic data and other geophysical surveys | Shipment data | |||
Mass balancing is inherently time-consuming and labor-intensive. Manual data collection, reconciliation and report generation consume valuable resources and time, leading to significant delays in identifying discrepancies and, more critically, the root causes of imbalances.
Sensors, despite their sophistication, are susceptible to errors, calibration issues and drift over time. The inherent lag between data collection and analysis significantly hinders proactive decision-making and the ability to respond rapidly to anomalies or potential operational issues. Moreover, the difficulty in accounting for unmeasured losses, such as fugitive emissions or evaporation, introduces significant blind spots into the mass balance equation.
Finally, the growing burden of compliance and reporting intensifies these challenges. Meeting stringent regulatory requirements for emissions and material accounting, along with the increasing need to provide verifiable data for sustainability certifications (e.g., mass balance for bio-based fuels), becomes an arduous and error-prone endeavor without advanced tools.
AI helps the oil and gas industry improve mass balance by providing better visibility into operations. Automating mass balance addresses the critical challenges above head-on. While automation provides the framework, integrating AI elevates mass balancing to an entirely new level of sophistication and effectiveness. Artificial intelligence is ideal to address the complex nature of material accounting within the oil and gas industry. AI will not replace existing mass balance software. Instead, it will supplement and enhance existing software.
In practice, the greatest value comes from hybrid modeling that couples data-driven AI/ML with mechanistic, first-principles process models. By grounding learning in conservation laws and unit-operation physics, hybrid approaches yield physically consistent estimates, improve extrapolation beyond seen conditions, and enhance interpretability. They can enforce constraints, use mechanistic models as features or priors, and adapt parameters online as plant conditions drift.
First, automation enhances accuracy. By eliminating human error and ensuring consistent data collection and processing, automated systems provide a more precise picture of material flows, leading to improved data validation and reconciliation. By continuously monitoring material flows, imbalances, anomalies, bottlenecks inefficiencies can be detected faster, minimizing material losses stemming from spills, leaks, or unaccounted-for product. Automated systems offer scalability and adaptability, easily integrating new sensors and data sources and adapting seamlessly to changing operational demands and process configurations, future-proofing the mass balancing infrastructure.
AI can ingest data from sensors and IoT devices, providing enhanced, granular data. Machine learning algorithms can then be employed to intelligently fuse data from disparate sources, creating a unified and coherent dataset for analysis. AI models are invaluable for data reconciliation and gap-filling. They intelligently reconcile conflicting measurements from various sensors, flow meters, tank levels, and other instruments, even learning from historical data to determine the most probable actual value, ensuring data accuracy, significantly improving the accuracy of material flow calculations even when data is incomplete.
Natural Language Processing (NLP) can unlock valuable insights from unstructured data sources. This capability includes extracting critical information from maintenance logs, shift reports and other text-based data, which, although not quantitative, can significantly impact mass balance understanding.
AI particularly excels in identifying anomalies and diagnosing faults. Machine learning algorithms identify even subtle deviations, flagging potential issues that human operators might miss. Predictive analytics can forecast potential imbalances before they fully develop, enabling proactive intervention and mitigation. Moreover, AI can effectively identify faulty sensors or instruments that are skewing data, enabling timely maintenance and calibration.
Looking to the future, the trajectory of mass balancing in the oil and gas industry is shifting toward greater sophistication and precision, with hybrid modeling emerging as the gold standard. AI systems are increasingly taking a proactive role in providing real-time insights, with greater integration anticipated with existing plant management systems.
The complex and critical nature of operations within the oil and gas industry demands an evolution in material flow management. Traditional mass balancing, with its inherent limitations and susceptibility to error, is increasingly inadequate in meeting today’s demands of efficiency, profitability and sustainability. The automation of mass balancing is no longer a luxury, but a critical necessity for enhanced accuracy, unlocking a future of operational excellence, increased sustainability and a significant competitive advantage in the global energy landscape.
Other articles that may interest you
