A hybrid ensemble-based automated deep learning approach to generate 3D geo-models and uncertainty analysis

Abstract

There is an increasing interest in creating high-resolution 3D subsurface geo-models using multisource retrieved data, i.e., borehole, geophysical techniques, geological maps, and rock properties, for emergency managements. However, dedicating meaningful, and thus interpretable 3D subsurface views from such integrated heterogeneous data requires developing a new methodology for convenient post-modeling analyses. To this end, in the current paper a hybrid ensemble-based automated deep learning approach for 3D modeling of subsurface geological bedrock using multisource data is proposed. The uncertainty then was quantified using a novel ensemble randomly automated deactivating process implanted on the jointed weight database. The applicability of the automated process in capturing the optimum topology is then validated by creating 3D subsurface geo-model using laser-scanned bedrock-level data from Sweden. In comparison with intelligent quantile regression and traditional geostatistical interpolation algorithms, the proposed hybrid approach showed higher accuracy for visualizing and post-analyzing the 3D subsurface model. Due to the use of integrated multi-source data, the approach presented here and the subsequently created 3D model can be a representative reconcile for geoengineering applications.

Normalizing Large Scale Sensor-Based MWD Data: An Automated Method toward A Unified Database

“…The database considers the implications of the research findings for practice and will help with consensus decisions on areas where evidence is not found. Accordingly, proper integration of such a unified database with geomechanical data can be the backbone of future deeper analyses through advanced computationally intelligent techniques [ 55 ]. Consequently, such databases offer more than just insights into the drilling; they also play a crucial role in optimizing the geoengineering operations and performance improvements via a reliable platform in terms of high-resolution 3D subsurface computer vision models based on the rock mass characteristics and geological mapping.…”

Mine Surface Settlement Prediction Based on Optimized VMD and Multi-Model Combination

“…To enhance predictive accuracy, it would be beneficial to incorporate additional parameters such as coal mine geology, rock characteristics, hydrogeology, development methods, and other relevant factors during the prediction process [49]. Further in-depth exploration and research are needed in future studies to reconstruct and model three-dimensional data [50].…”

Review of driving-behaviour simulation: VISSIM and artificial intelligence approach

“…Two factors influenced this issue: the magnitude of the ANN and the length of time dedicated to training the ANN. Overfitting pertains to a situation in which the size of an ANN surpasses its ideal capacity, whereas overtraining refers to the period of training an ANN that could ultimately lead to a decrease in the network's predictive capability [ 162 , 163 ]. Overfitting is a phenomenon that arises when a model acquires an excessive amount of intricate information, including noise from the training dataset.…”