Paul Baville

Thesis topic:  Complexe sedimentary basin models uncertainties assessment by simulating stochastically well correlations from log and core data using the Dynamic Time Warping algorithm.

Thesis supervisor: Guillaume Caumon (Univ. Lorraine - RING Team)

Thesis co-supervisors: Cédric Carpentier (Univ. Lorraine - CNRS) & Marcus Apel (Equinor)

Thesis monitoring committee:  Antoine Tabbone (Univ. Lorraine - Loria) & Dirk Knaust (Equinor)

Research Topics:

  • Stochastic Simulation of Stratigraphic Sequences for Wells.
    • Maximum Flooding Surfaces (MFS), which correspond to most marine depositional conditions, may be interpreted from Gamma Ray. Indeed, high value of Gamma Ray might correspond to most marine depositional conditions. Using Discrete Wavelet Transform (DWT) applied on gamma ray log, we can compute a normalized cumulative density function of finding a MFS at each depth. Using this normalized cdf, MFS may be stochastically simulated.
  • Computer-assisted chronostratigraphic multi-well correlation.
    • Well correlation based on facies interpretations and paleo-distality. Markers are associated if they are consistent with depositional environments. Along the structural strike direction, chronostratigraphy tends to associate similar facies. At the opposite, along the structural dip direction, proximal shallow facies are associated with distal deep facies, other associations are exculded.
    • Well correlation based on facies interpretations and dipmeter data. Chronostratigraphic surfaces are simulated between well markers using Bézier cubic interpolations. Every possible interpolations are compared with a theoretical depositional profile and the smaller the variation, the higher the likelihood.
  • Stochastic dipmeter data computation from four-directional resistivity log correlation.
    • Using the Dynamic Time Warping (DTW) to correlate four-directional resistivity logs, it is possible to associate resistivity markers minimizing the variance. Once markers associated, it is possible to compute dips and dip directions for each marker associations.
  • Bayesian updating of horizontal well trajectory while drilling using relative geological time.
  • Magnetostratigraphic time-to-depth multi-well correlation

Achievements (2019 - 2020):

  • Computer-assisted multi-well stochastic correlations: Chronostratigraphic surface simulation using Bézier cubic interpolations
  • Magnetostratigraphic time-to-depth multi-well correlation using the Dynamic Time Warping on well-log data.
  • Facies correlation based on lithostratigraphic and chronostratigraphic associations based on the paleo-environment (transport direction, paleo-depth, paleo-distality ...) using Dynamic Time Warping on well-log data.

Achievements (2018 - 2019):

  • Stochastic detection of stratigraphic sequences from well data (Gamma Ray log) using discrete wavelet transform (Haar functions).
  • Sedimentological interpretation of well core samples and creation of different well logs based on these different interpretations (Petrophysical facies, genetic facies …) + parasequence interpretations.
  • Facies transition probability computation to constraint stochastic correlations using observed data.
  • Stochastic multi-well correlation from well data (Gamma Ray log + genetic facies interpretation on core samples) using Dynamic Time Warping algorithm to compute the n-best correlations.

Contact Information

E-Mail :This email address is being protected from spambots. You need JavaScript enabled to view it.
Phone number:+33 3 72 74 45 19
Cell:+33 6 67 67 09 58
ENSG office number:F221