This online training course will take an introductory look at the opportunities, challenges and specific requirements for petroleum data analytics for the energy industry.

​Real-time data now represents a growing stream of scores of channels of digital data being fed concurrently to numerous receiving entities in the operator's remote monitoring amenities, at contractor centers and other facilities. Analytics are applied to data channels to signal or predict deviations from expected readings that require attention. For such systems to work effectively and reliably the input data must be trustworthy.

This presentation describes the development of a well-site automation system consisting of an open data aggregator, with networked surface and downhole sensors and real-time applications for process monitoring, advice and control.

Analytics is not new to the oil & gas industry. From the early days of seismic acquisition and processing, well log interpretation and reservoir simulation, industry experts have used various techniques up to the limit of existing computer power, to analyze and model data recorded in the field to make better decisions.

Data volume is exploding — more than 90% of today’s data was created in the last few years, and there is an exponential increase in new types of data, with mobile, social media, video, and the Industrial Internet of Things adding to the growth of seismic, reservoir, drilling, production, and engineering data.

Presented by Dr. Luigi Saputelli

This presentation will strive to give an overview of the key components to build a fully automated system to identify rig states and provide rig operations performance key performance indicators (KPIs). Starting with the data requirement, feature engineering, machine learning, and how the sensor’s noise can be reduced in real-time. It will also provide information on drilling operation’s specific challenges (such as KPI validation) and how it can be addressed. Finally, we will discuss lessons learned of running such a system, and how such an effort can establish the foundation of advanced data science driven solutions.

The purpose of this seminar is to illustrate the effectiveness of Computational Fluid Dynamics (CFD) in guiding re-design of a scrubber inlet within the constraints of the project team.

Digital Oilfield is an Intelligent Energy concept that ensures a continuous optimisation of an asset or group of assets through integration of standard tools, right people, defined business processes and appropriate facility or space.

This is group of webinars covers various topics on Data-Driven Analytics.

This intermediate-level course will explain some of these measures and tools, describe some specific measurements, and explain the relationship between assessment and strategy. Summarizing the data with the correct tool can be the gating factor to reaching staff and effecting changes that spur performance improvement.

This intermediate-level course will provide answers to these questions as well as recommendations for decision-making based on data analytics for each sector. The course will begin with an introduction of Big Data, then continue into a deeper dive on its implications within each sector. Industry case studies make the concepts applicable in the real-world.

The concepts behind the digital oilfield will be discussed and how they fit together into creating an ecosystem to deliver value. Concrete elements will be presented that illustrate the entire value chain and technology stack necessary to deliver on the promise of the digital oilfield. We illustrate this vision with a concrete example in which live data from an entire oilfield full of sucker-rod pumps is fed into a central facility, analyzed for predictive maintenance, and used to perform proactive maintenance on the pumps.

Data science projects exhibit many challenges that drive most of these projects into failure. Data scientists should equip themselves with different techniques: coding, business skills, data handling skills in order to deliver a successful project. As it is reported, data scientists spend 80% of their time on cleaning and preparing data. Thus, a high focus should be devoted to this step in order to be done the right way.

Deploying machine learning-based predictive models to the oil field is quite challenging. They are remote, hazardous, and have spotty connectivity to the cloud. The world of operationalizing a model is very different than the perfect lab environment where the models are born.