A Study of Compression Methods and Their Applications to Large Geological Models
Abstract
The promise of deep learning is to discover rich hierarchical structure of the data that allows to circumvent manual feature extraction and engineering. In this study, we explore novel applications of the deep learning, building on the representational power of the neural networks to find latent variables governing the reservoir model. In Chapter I, we first introduce the relevant petroleum engineering background and motivation. Then, we give an overview of the state-of-the-art traditional and deep learning machine learning models explored in the previous works. After extensively covering the pitfalls in the traditional modeling approaches, we propose our model to significantly improve the performance. We conclude with Chapter II where we discuss possible uses of the latent reservoir representation, e.g., decline curve prediction.
Subject
compressiondeep learning
machine learning
geo-model
reservoir engineering
petroleum engineering
decline curve prediction
Citation
Churilova, Polina D. (2023). A Study of Compression Methods and Their Applications to Large Geological Models. Master's thesis, Texas A&M University. Available electronically from https : / /hdl .handle .net /1969 .1 /199011.