Sign up for our newsletter and get the latest HPC news and analysis.
Send me information from insideHPC:


Energy Companies Embrace Deep Learning for Inspections, Exploration & More

This is the second entry in a five-part insideHPC series that takes an in-depth look at how machine learning, deep learning and AI are being used in the energy industry. Read on to learn how energy companies are embracing deep learning for inspections, exploration and more. 

deep learning

Download the full report.

Deep Learning

Simulation has been used for many years to understand where deposits may lie underground. The amount of data that may be analyzed is growing in order to make more precise decisions on where the most likely location is to drill. Alone, an increasing number of simulations is not enough to increase location productivity.

Royal Dutch Shell has embraced AI to assist customers in searching for the right lubricant for the needed project.

This is an example of where deep learning using neural networks can play a large role in increasing the understanding of what is in the ground below. Artificial neural networks, which involve large numbers of computing elements, such as those found in today’s GPUs, take complex data and compare it to previous patterns to assist in making more accurate decisions.

Total uses AI with autonomous robots that carry out inspections, currently performed by humans, to detect abnormal equipment activity and intervene in an emergency. Examples may include simple equipment malfunctions, or more high-risk situations such as gas leaks.

The need for simulation based on the current understanding of the physical world remains as strong as ever, but can now be combined with algorithms that have been trained to recognize certain data models. By combining different technologies, the sum becomes greater than the individual parts. Incorporating previously learned knowledge automatically into the discovery process leads to decreased exploration costs.

deep learning

Figure 1 – Opportunities for machine learning as part of a workflow

While simulations that take advantage of thousands of processors can deal with data that has already been preprocessed, there is an increasing need to process the tremendous amounts of data produced before the simulation even begins. Streaming data must be ingested and turned into meaningful data that can be used. Figure 1 above shows various stages of the exploration process where machine learning can be integrated into a workflow.

Over the next few weeks, this insideHPC Guide series will also cover the following topics:

Download the full report, “Machine Learning in Energy: A Hot Spot in Seismic Processing,” courtesy of Dell EMC and Nvidia, to learn more about how machine learning is being used to move the energy industry forward. 

Leave a Comment

*

Resource Links: