2021 SPWLA PDDA Machine Learning Contest - Call for Registration
SPWLA PDDA SIG is excited to announce its 2021 machine learning contest!
This contest is open to all petrophysics enthusiasts. Top winning teams will be awarded prizes and invited to present at the PDDA SIG annual meeting. Please register with Lei Fu ([email protected]) by submitting your team information (names, affiliations, and contacts), before October 15, 2021. The dataset comes from the Equinor Volve field. A data repository is available at: https://github.com/pddasig/Machine-Learning-Competition-2021
Sponsoring Opportunities: SPWLA PDDA SIG is accepting sponsorship for this event to award the top winning teams. Please contact Lei Fu ([email protected]) for details.
SPWLA PDDA SIG Contest Committee: Lei Fu, Chicheng Xu, Yanxiang Yu, Michael Ashby, McDonald Andy, Bin Dai
Task: Well-log based reservoir property estimation with machine learning
Background: Well logs are interpreted/processed to estimate the in-situ reservoir properties (petrophysical, geomechanical, and geochemical), which is essential for reservoir modeling, reserve estimation, and production forecasting. The modeling is often based on multi-mineral physics or empirical formulae. When sufficient amount of training data is available, machine learning solution provides an alternative approach to estimate those reservoir properties based on well log data and is usually with less turn-around time and human involvements.
Problem Statement: The goal of this contest is to develop data-driven models to estimate reservoir properties including shale volume, porosity, and fluid saturation, based on a common set of well logs including gamma ray, bulk density, neutron porosity, resistivity, and sonic. You will be provided with log data from about 10 wells from the same field together with the corresponding reservoir properties estimated by petrophysicists. You need to build a data-driven model using the provided training dataset. Following that, you will deploy the newly developed data-driven models on the test dataset to predict the reservoir properties based on the well log data.
- Contestant can be an individual or a group with the maximum size of 4.
- The contest focuses on data-driven methods, the use of additional data or petrophysical equations is not allowed.
- Privately sharing code or data outside of teams is not permitted. However, it's okay to share code if made available to all participants on the competition Github repository via submitting issues or pull requests.
- A contestant will submit the estimated reservoir properties for testing wells as separate .csv files.
- A contestant will submit the source code and a brief report documenting the accuracy achieved in a few plots.
- The judges will review the source code.
- The performance of the model will be quantified in terms of root mean square error (RMSE).
- A leaderboard will be updating the rank of submissions from each team.
- The contestant with the best quality source code and the best performance will be declared the winner for this competition.
SPWLA Petrophysical Data-Driven Analytics Special Interest Group (PDDA-SIG)
Following the positive feedback of the recently conducted special interest group (SIG) survey during 2018 SPWAL Spring topical conference, we are pleased to announce the formation of a new SPWLA special interest group dedicated on Petrophysical Data-Driven Analytics (PDDA). The vision for establishing a PDDA-SIG began at the 2018 SPWLA Spring topic conference in Houston, where more than 60 industry professionals (core analyst, tool physicists, petrophysists, geologist etc.) and data scientists attended to discuss the applications of advanced data analytics techniques to challenging petrophysical interpretations and oil field operations. Most of the conference attendees showed great interest and preregistered themselves as members of SPWLA PDDA SIG.
The goals of the PDDA-SIG are to create a venue for exchanging and sharing knowledge and best practices of applying advanced data analytics to solve challenging big-data-related problems in the oil and gas exploration and development. It aims to foster the cross-disciplinary technical collaborations between practitioners in academic and different sectors of O&G industry, and to promote networking for industry professionals.
The PDDA-SIG will cover the following primary technical areas:
sensor and logging technology
rock physics and geomechanics
oil field operation improvement
data QC and management
AI assisted automation
Participation in the PDDA-SIG is open for any practitioner and student who are interested in this subject matter. Please use following URL in LinkedIn Group for PDDA-SIG member registration:
The PDDA-SIG executive committee (2021-2022) is composed of:
Chairman: Yanxiang Yu, Shell, [email protected]
Vice Chairman: Andy McDonald, Lloyd's Register, [email protected]
Secretary of Publications: Bin Dai, Halliburton, [email protected]
Treasurer: Lei Fu, Aramco, [email protected]
Advisor: Michael Ashby, Devon Petroleum Corporation, [email protected]
Advisor: Chicheng Xu, Aramco, [email protected]
From 2019 to 2020, PDDA organized a machine learning competition, please see the detail here:
GitHub - pddasig/Machine-Learning-Competition-2020: SPWLA PDDA’s 1st Petrophysical Data-Driven Analytics Contest -- Sonic Log Synthesis
PDDA is now calling for proposals for Machine learning 2021 competition, please contact Lei Fu if you like to submit a proposal or sponsor the event.