Fully Unmanned Operations Controlled from Centralised Operations Centers Operating Autonomous Including Fully Automated Response to Disturbances

Fully Unmanned Operations Controlled from Centralised Operations Centers Operating Autonomous Including Fully Automated Response to Disturbances

06 Sep 2018, 13:45 - 15:15

Crombie B
Language:
English

Session Title: Fully Unmanned Operations Controlled from Centralised Operations Centers Operating Autonomous Including Fully Automated Response to Disturbances 
     

1345-1415: Presentation: A digital twin concept and how it will help to shape the future of autonomous operations.

Machine learning algorithms trained by a simulation model can learn how to react to unexpected situations, which can augment the safety of remote or autonomous operation of assets. With a specific digital twin concept, it’s possible to combine multiphase models for flow protection, process, maritime, and many other component models in a large dynamic process. This gives the user a 3D interface to the platform where they can access, monitor and predict the outcome of processes that over time can be made more autonomous.                                                                                           

Speaker: Andreas Jagtøyen, Kongsberg
                                                   

1415-1445: Accidents are always bound to happen. However, cutting edge technologies such as natural language processing (NLP) powered by machine learning can help oil and gas technicians generate insights from collected data, detect irregular patterns, and minimize safety issues in operations and maintenance processes. NLP can be used for query expansion, data classification, and information discovery to improve safety in drilling operations by using a wealth of information unable to be leveraged by humans alone. The use of cognitive algorithms for information retrieval, NLP classification to improve the quality of data, and word-embedding technology to surface hidden insights in safety records helps to prevent industrial injuries by understanding near-misses likely to recur and events that can be avoided.

Philippe Herve, SparkCognition
                                   

1445-1515: Discussion

We use cookies to operate this website and to improve its usability. Full details of what cookies are, why we use them and how you can manage them can be found by reading our Privacy & Cookies page. Please note that by using this site you are consenting to the use of cookies.