Big Data in Subsea Solutions Subsea Valley Conference 2014 Telenor Arena, Fornebu, April 2-3 Roar Fjellheim, Computas AS
Computas AS - Brief company profile Norwegian IT consulting company providing services and solutions for work processes, business systems and knowledge-based collaboration 280 highly qualified project managers, software architects and developers, and business consultants Large customers in government and private sector, including the oil & gas industry 1985 spinoff from DNV, Høvik Located at Lysaker, Stavanger and Romania www.computas.com 2
Computas - Solutions and services Business process management System architecture and integration Software engineering services Integrated operations (CODIO) Compliance solutions (UCMS ) Big Data - Information Management Knowledge management services Consulting and project management 3
Outline What is Big Data? Big Data in Oil & Gas Subsea Applications Big Data Enablers What s next? 4
What is Big Data? Gartner s definition: Big Data is high-volume, - velocity and -variety information assets demanding innovative forms of information processing for enhanced insight and decision making. 5
Big Data Drivers and enablers The world now creates 5 Exabyte (10 6 Terabyte) per 2 days (= all data created from dawn of civilization to 2003) The Internet and WWW 1/3 of world s population online The Internet of Things (IoT) Sensors everywhere How will society and industries be able to manage and benefit from this avalanche of information? Rapid advances in digital technologies Smart algorithms and intelligent machines New business models Oil & gas including subsea is riding this wave! 6
Digitalization of field operations Integration of people, process and technology for faster and better decisions, based on real-time data and integrated work processes Smart sensors, downhole etc. High bandwidth networks Simulators and models Advanced optimization Data visualization Higher levels of automation Offshore/onshore workflows Source: NTNU IO Center 7
Subsea challenges and Big Data opportunities The usual suspects : Longer, deeper, colder, HSE, reliability, cost and efficiency ever more important Increasingly sophisticated machinery deployed subsea Big Data opportunities: Condition Based Monitoring and Maintenance Integrated Production Optimization Enhanced Logistics Support And more 8
Condition based maintenance (CBM) Technical integrity - Strict operational and regulatory requirements Inspection, maintenance and repair - major OPEX drivers CBM requirements Continuous analysis and diagnostic of equipment sensor data (descriptive analytics) Model-based prediction of probable failures and remaining equipment life time (predictive analytics) Planning (re-planning) and optimization of inspection and intervention activities (decision analytics) New business models for operator supplier relationships, cf. how the auto industry uses remote car monitoring 9
Example - Router module leakage detection Condition indicator Pressure B Normal 3 months from CPM detection to automatic switch over Cause: Fault in fiber penetrator in router module Effect: Leakage into module leading to communication failure Mitigation: CPM identified increased internal pressure due to leakage 3 months prior to communication issues Action: Router module changed out in planned intervention campaign Source: FMC Technologies 10
Big Data enabler - Data Science How to extract knowledge from data in order to better understand, predict and decide Predictive modeling Event classification Semantic analysis Data mining Visualization Machine learning Decision optimization Smart Algorithms! 11
Example - Decision network for drilling Based on data from sensors and other sources Predicts and simulates consequences of different decisions Recommends action with highest expected value 12 Source: ConocoPhillips, Computas 12
Integrated production optimization Daily well-related decisions to optimize production Goal: Fully utilize production and process system capacity Multi-well, cannot optimize single wells in isolation Constrained by Reservoir properties (changing) Technical layout and capacities Enablers Increased instrumentation and data rates allows better models Improved control mechanisms - Choking, lifting, routing Result More frequent, optimized control Integrated optimization of overall multi-well installation 13
Supervised closed-loop production optimization Source: NTNU IO Center 14
Big Data enabler New IT frameworks Big Data Volume and Velocity Parallel computation MapReduce Hadoop Real time data processing In-memory data management Stream processing Intelligence engines Complex event processing (CEP) Online multivariate analysis 15
Enhanced logistics support ELH - EPIM Logistics Hub A common platform for operators and suppliers to exchange logistics events, e.g. container movements EPIM is a association of companies operating on the Norwegian Continental Shelf, creates common solutions with a 5x advantage over individual solutions Supplier The objective is to trace all containers going in and out to facilities on the NCS, improving efficiency, Supply Base reliability and economy Operator of logistic movements Rig Owner Transporter 16
Big Data enabler Semantic technology Big Data Variety Structured and «unstructured» data Many different data types and data sources Large scale Information Management Relational view of data Simplifies complex data models Supports meta-data management Enables data integration Based on open, international standards W3C Semantic Web, RDF, ISO 15926, etc. 17
A look ahead - Autonomous subsea systems Future subsea systems will be required to operate for extended periods of time without human guidance Autonomy, the ability to to self-diagnose and repair, and to plan (and re-plan) actions based on high-level goals, using Big Data and related technologies Autonomy already exists in selected areas, like ROVs for pipeline inspection, but will have much wider impact Ref. «Autonomous Systems for the Oil & Gas Industry», NFA - Norwegian Society of Automatic Control, 2013 18
Main takeaways Big Data is a reality (even if hyped) 3Vs Volume, Velocity and Variety Many oil & gas applications Data Science and new IT frameworks From data to insight, prediction and decisions Thank you for your attention! 19