Poster Open Access
Combining Physics-Based and Data-Driven Modeling for Pressure Prediction in Well Construction
Oney Erge;
Eric van Oort
This poster introduces a novel framework to combine the physics-based and data-driven modeling, aiming to attain the best features of both approaches for well construction. Gaussian processes, neural networks and deep learning models are trained and executed together with a physics model that is directly derived using the first principles. Then the results are combined through a decision-making algorithm, a hidden Markov model. The approach is tested within the scope of wellbore hydraulics on a dataset from an actual drilling operation. The results suggest the proposed approach has a good potential to allow safer, optimized drilling operations.
| Name | Size | |
|---|---|---|
|
SciPy2020_poster.pdf
md5:5bc9c2dc964c594d9cab8368b19cd0ff |
1.0 MB | Download |
| All versions | This version | |
|---|---|---|
| Views | 219 | 219 |
| Downloads | 201 | 201 |
| Data volume | 209.5 MB | 209.5 MB |
| Unique views | 205 | 205 |
| Unique downloads | 182 | 182 |
- Publication date:
- June 24, 2020
- DOI:
-
- Keyword(s):
- Deep Learning, Machine Learning, Combining Physics-Based Modeling and Data-Driven Modeling, Hydraulics Modeling, Frictional Pressure Loss Modeling.
- License (for files):
- Creative Commons Attribution 4.0 International