Subject ▸ Volve

A simple Binder project with a Volve dataset

Test binder with simple Volve project. Using the project set up by Navarro on well logs.

Build with R and Jupyter.

A Vagrant virtual machine to run data science on Volve datasets

vagrant-volve-navarro-BI64G20S2JP8201 This is reproducible work of Machine Learning and Data Science applied to data from the Volve field. Features This is a VirtualBox Virtual Machine (VM) that is automatically generated using Vagrant. A few Machine Learning and Deep Learning packages have been installed, such as Scikit-Learn, NLTK, Keras, TensorFlow and Theano. A Vagrant file is used to generate this VM based on Ubuntu 18.04 (bionic64). Additional packages required for this phase of the ML and DS work are welly, pandas, numpy, seaborn, and lasio.

Read More…

The impact of the Volve dataset

This is a part of the response I wrote to a reader in my [blog]( The question was about the limited scope of the Volve license, questioning its openness, because it doesn’t cover commercialization from the data. This was my response: I am not finding any problem whatsoever. For my purposes of research, learning, education and teaching, I think the license is alright. In all articles and papers I always give the corresponding attribution.

Read More…

Integrating Python and R for data science. Converting Eclipse binary files to dataframes in the Volve dataset

Introduction Python and R offer a good combination of powers: dozens of proven engineering, data science, and machine learning libraries, also a science oriented approach towards full reproducibility. As I have told you before, I started my coding journey with Python many years ago. I even wrote a large application for production optimization using OpenServer, Prosper, GAP and MBAL by Petroleum Experts while I was on my 3-year tour with Petronas in Kuala Lumpur in Malaysia.

Read More…

Volve dataset. First look of the Reservoir simulation models.

The Eclipse reservoir models from the Volve dataset working like a charm. The compressed file is 399 MB in size.

I was able to open the models with ResInsight (thank you Matthew Kirkman). The software is open source and relatively easy to use.

Here is the Eclipse case opened.


Volve datasets: what can we learn?

Let’s enter the individual size of each of the Volve data files: # megabytes cat(sprintf("Megabytes\n")) #> Megabytes (mega <- c(geophysics = 99, reservoir_eclipse = 390, well_technical = 212, seismic_vsp = 95, production = 2, reports = 162)) #> geophysics reservoir_eclipse well_technical seismic_vsp #> 99 390 212 95 #> production reports #> 2 162 # gigabytes cat(sprintf("Gigabytes\n")) #> Gigabytes (giga <- c(geoscience = 54.6, reservoir_rms = 2.1, well_logs = 6.

Read More…