scholarly journals Data-driven optimization of brittleness index for hydraulic fracturing

2021 ◽  
Author(s):  
lei hou ◽  
Jianhua Ren ◽  
Yi Fang ◽  
Yiyan Cheng
Keyword(s):  
Machine Learning ◽  
Hydraulic Fracturing ◽  
Fundamental Principle ◽  
Data Driven ◽  
Brittleness Index ◽  
Backward Elimination ◽  
Promising Application ◽  
Independent Variable ◽  
Using Data ◽  
New Criterion

Evaluation of brittleness index (BI) is a fundamental principle of a hydraulic fracturing design. A wide variety of BI calculations often baffle field engineers. The traditional value comparison may also not make the best of BI. Moreover, it is often mixed up with the fracability in field applications, thus causing concerns. We, therefore, redefine fracability as the fracturing pressure under certain rock mechanical (mainly brittleness), geological and injecting conditions to clarify the confusion. Then, we propose a data-driven workflow to optimize BIs by controlling the geological and injecting conditions. The machine learning (ML) workflow is employed to predict the fracability (fracturing pressure) based on field measurement. Three representative ML algorithms are applied to average the prediction, aiming to restrict the interference of algorithm performances. The contribution of brittleness on pressure/fracability prediction by error analysis (rather than the traditional method of BI-value comparison) is proposed as the new criterion for optimization. Six classic BI correlations (mineral-, logging- and elastic-based) are evaluated, three of which are optimized for the derivation of a new BI using the backward elimination strategy. The stress ratio (ratio of minimum and maximum horizontal principal stress), representing the geological feature, is introduced into the derived calculation based on the independent variable analysis. The reliability of the new BI is verified by error analyses using data of eight fracturing stages from seven different wells. Approximately 40%~50% of the errors are reduced based on the new BI. The differences among the performances of algorithms are also significantly restrained. The new brittleness index provides a more reliable option for evaluating the brittleness and fracability of the fracturing formation. The machine learning workflow also proposes a promising application scenario of the BI for hydraulic fracturing, which makes more efficient and broader usages of the BI compared with the traditional value comparison.

scholarly journals Predicting the state of charge and health of batteries using data-driven machine learning

2020 ◽  
Vol 2 (3) ◽  
pp. 161-170 ◽  
Author(s):  
Man-Fai Ng ◽  
Jin Zhao ◽  
Qingyu Yan ◽  
Gareth J. Conduit ◽  
Zhi Wei Seh
Keyword(s):  
Machine Learning ◽  
The State ◽  
State Of Charge ◽  
Data Driven ◽  
Using Data

scholarly journals AERIAL PHOTOGRAMMETRY AND MACHINE LEARNING BASED REGIONAL LANDSLIDE SUSCEPTIBILITY ASSESSMENT FOR AN EARTHQUAKE PRONE AREA IN TURKEY

2021 ◽  
Vol XLIII-B3-2021 ◽  
pp. 713-720
Author(s):  
G. Karakas ◽  
S. Kocaman ◽  
C. Gokceoglu
Keyword(s):  
Machine Learning ◽  
High Resolution ◽  
Landslide Susceptibility ◽  
Data Driven ◽  
Natural Phenomenon ◽  
Landslide Occurrence ◽  
Conditioning Factors ◽  
Aerial Photogrammetry ◽  
Prone Area ◽  
Using Data

Abstract. Landslide is a frequently observed natural phenomenon and a geohazard with destructive effects on economies, society and the environment. Production of up-to-date landslide susceptibility (LS) maps is an essential process for landslide hazard mitigation. Obtaining up-to-date and accurate data for the production of LS maps is also important and this task can be achieved by using aerial photogrammetric techniques, which can produce geospatial data with high resolution. The produced geospatial datasets can be integrated in data-driven methods for obtaining accurate LS maps. In the present study, LS map was produced by using data-driven machine learning (ML) methods, i.e. random forest (RF). An earthquake and landslide prone area from the south-eastern part of Turkey was selected as the study area. Topographical derivatives were extracted from digital surface models (DSMs) produced by using aerial photogrammetric datasets with 30 cm ground sampling distances. The lithological parameters were employed in the study together with an accurate landslide inventory, which were also delineated by using the high-resolution DSMs and orthophotos. The relationships between the landslide occurrence and the pre-defined conditioning factors were analyzed using the frequency ratio (FR) method. The results show that the RF method exhibits high prediction performance in the study area with an area under curve (AUC) value of 0.92.

scholarly journals AI oriented prospectivity mapping to study relationships between Sb mineralization and geological framework

2021 ◽  
Author(s):  
Alex Vella ◽  
Charles Gumiaux ◽  
Guillaume Bertrand ◽  
Bruno Tourlière ◽  
Eric Gloaguen ◽  
...  
Keyword(s):  
Machine Learning ◽  
Multiple Scales ◽  
Mineral Exploration ◽  
Data Driven ◽  
Automated Classification ◽  
Geological Features ◽  
Data Driven Approach ◽  
Statistical Relationships ◽  
Using Data ◽  
Prospectivity Mapping

<p>Prospectivity mapping aims at producing favorability maps, outlining areas with the highest likelihood to host mineralization. This process can be done using data-driven approaches, based on statistical and spatial analyses on geological features and known mineral occurences. Besides, such approach contributes to better understand metallogenic processes by highlighting specific and systematic associations between deposits and geological features (structures, lithologies, contacts, geophysical anomalies, etc).</p><p>As part of the AUREOLE project, prospectivity maps of Sb throughout the West European Variscan Range are being produced using CBA (“Cell-Based Associations”). CBA is a prospectivity tool developped for mineral prospectivity mapping by the French Geological Survey (BRGM). This method divide at first space into a regular cells grid. Inside each cell, the associations of geological factors, such as lithological, structural, geophysical or geochemical features, are grouped together and define the geological framework in the vicinity of the given cell. This project aims at developing and improving this method by the addition of new machine learning methods and statistical and spatial analysis tools for the automated classification and the calculation of favorability score.</p><p>Application of this approach to the Ibero-Armorican Arc, relying heavily on Artificial Intelligence to process the data, will highlight statistical relationships between the Sb deposits and their surrounding geological framework. Computations will be performed at multiple scales and in different areas trough the Arc, in order to observe the influence of scale in the consistency of the results and to bring out general laws from local specificities in the metallogenic models. Results from this almost purely data-driven approach will be compared to the metallogenical models traditionnaly proposed for Sb deposits in the studied areas. We infer this new multiscale and multidomains study  would improve our understanding of the genetic processes resulting in  Sb deposits through the Variscan Range and give new metallotects or specify the common ones, to be used for mineral exploration purpose.</p><p>This Phd work is funded by the ERA-MIN2 AUREOLE project (ANR-19-MIN2-0002, https://aureole.brgm.fr).</p><p><strong>Keywords: </strong>Antimony, Prospective Mapping, Machine Learning, Data-Driven.</p>

Sign in / Sign up

Export Citation Format

Share Document